技术领域Technical Field
本公开涉及终端技术领域，尤其涉及一种唤醒控制方法、装置、存储介质及终端。The present disclosure relates to the field of terminal technology, and in particular to a wake-up control method, device, storage medium and terminal.
背景技术Background technique
随着科技的发展，越来越多的智能设备逐渐走进用户的生活，在智能设备中，语音控制、语音输入以及语音启动等应用变得越来越受欢迎。智能设备通过搭载语音识别系统，可以实时采集用户的语音数据，并根据该语音数据执行用户发送的控制指令，与用户进行交互。With the development of science and technology, more and more smart devices are gradually entering the lives of users. In smart devices, applications such as voice control, voice input and voice activation are becoming more and more popular. Smart devices can collect user voice data in real time by being equipped with a voice recognition system, and execute control commands sent by users based on the voice data to interact with users.
但是，在现实环境中，由于噪声的干扰，根据实时语音数据响应用户的控制指令时，出现错误识别的概率比较高，使得现行的语音识别系统的准确率较低。However, in a real environment, due to noise interference, the probability of misrecognition when responding to user control commands based on real-time voice data is relatively high, resulting in a low accuracy rate of current voice recognition systems.
发明内容Summary of the invention
为克服相关技术中存在的问题，本公开提供一种唤醒控制方法、装置、存储介质及终端。In order to overcome the problems existing in the related art, the present disclosure provides a wake-up control method, device, storage medium and terminal.
根据本公开实施例的第一方面，提供一种唤醒控制方法，所述方法包括：采集多路音频数据；分别对目标时间段内采集的每路第一音频数据进行信号处理，得到多个第一目标音频数据；分别获取多个所述第一目标音频数据的第一置信度，其中，所述置信度用于表征所述音频数据能够唤醒终端的概率；获取历史时间段内第二目标音频数据的第二置信度，所述历史时间段为所述目标时间段之前的预设时间段，所述第二置信度包括多个所述第二目标音频数据的置信度；根据所述第一置信度和所述第二置信度确定是否唤醒所述终端。According to a first aspect of an embodiment of the present disclosure, a wake-up control method is provided, the method comprising: collecting multiple channels of audio data; performing signal processing on each channel of first audio data collected within a target time period to obtain multiple first target audio data; obtaining first confidences of the multiple first target audio data respectively, wherein the confidences are used to characterize the probability that the audio data can wake up the terminal; obtaining second confidences of second target audio data within a historical time period, the historical time period being a preset time period before the target time period, the second confidence comprising confidences of multiple second target audio data; and determining whether to wake up the terminal based on the first confidence and the second confidence.
可选地，所述分别对目标时间段内采集的每路第一音频数据进行信号处理，得到多个第一目标音频数据包括：选取所述终端的麦克风阵列的一个麦克风作为参考通道；获取所述参考通道在所述目标时间段采集的参考音频数据；根据所述参考音频数据，通过多个信号处理方式分别对每路所述第一音频数据进行信号处理，得到多个所述第一目标音频数据，不同路的第一音频数据的信号处理方式不同。Optionally, the signal processing of each channel of the first audio data collected within the target time period to obtain multiple first target audio data includes: selecting a microphone of the microphone array of the terminal as a reference channel; obtaining the reference audio data collected by the reference channel during the target time period; and according to the reference audio data, performing signal processing on each channel of the first audio data by multiple signal processing methods to obtain multiple first target audio data, and the signal processing methods of different channels of first audio data are different.
可选地，所述分别获取多个所述第一目标音频数据的第一置信度包括：针对多个所述第一目标音频数据中的每个所述第一目标音频数据，确定该第一目标音频数据对应的信号处理方式；根据该信号处理方式确定该第一目标音频数据对应的目标解码器，其中，不同的信号处理方式对应不同的解码器；将该第一目标音频数据输入所述目标解码器进行解码处理，得到该第一目标音频数据的第一置信度。Optionally, respectively obtaining the first confidence of multiple first target audio data includes: determining, for each of the multiple first target audio data, a signal processing method corresponding to the first target audio data; determining a target decoder corresponding to the first target audio data according to the signal processing method, wherein different signal processing methods correspond to different decoders; and inputting the first target audio data into the target decoder for decoding processing to obtain the first confidence of the first target audio data.
可选地，所述根据该信号处理方式确定该第一目标音频数据对应的目标解码器包括：通过预先设置的解码器关联关系，从多个所述解码器中确定该信号处理方式对应的解码器，所述解码器关联关系包括不同的信号处理方式和所述解码器的对应关系；将该信号处理方式对应的解码器作为所述目标解码器。Optionally, determining the target decoder corresponding to the first target audio data according to the signal processing method includes: determining a decoder corresponding to the signal processing method from a plurality of decoders through a pre-set decoder association relationship, the decoder association relationship including a correspondence between different signal processing methods and the decoders; and using the decoder corresponding to the signal processing method as the target decoder.
可选地，所述根据所述第一置信度和所述第二置信度确定是否唤醒所述终端包括：在每获取到一个所述目标解码器输出的所述第一置信度的情况下，根据所述目标解码器输出的第一目标置信度执行以下唤醒处理方式，直至唤醒所述终端，或者，根据多个所述目标解码器输出的所述第一置信度均执行了所述唤醒处理方式；所述唤醒处理方式包括：根据所述第二置信度和所述目标解码器输出的第一置信度确定是否唤醒所述终端。Optionally, determining whether to wake up the terminal based on the first confidence and the second confidence includes: each time the first confidence output by the target decoder is obtained, performing the following wake-up processing method according to the first target confidence output by the target decoder until the terminal is woken up, or performing the wake-up processing method according to the first confidence output by multiple target decoders; the wake-up processing method includes: determining whether to wake up the terminal based on the second confidence and the first confidence output by the target decoder.
可选地，所述根据所述第一置信度和所述第二置信度确定是否唤醒所述终端包括：从所述第二置信度中确定目标置信度，所述目标置信度和所述第一置信度是通过相同解码器解码得到的置信度；根据所述目标置信度和第三置信度，获取所述第一置信度对应的权重值，所述第三置信度包括所述第二置信度中除所述目标置信度之外的其他置信度；根据所述权重值和所述第一置信度，确定最终置信度；根据所述最终置信度确定是否唤醒所述终端。Optionally, determining whether to wake up the terminal based on the first confidence and the second confidence includes: determining a target confidence from the second confidence, the target confidence and the first confidence are confidences obtained by decoding by the same decoder; obtaining a weight value corresponding to the first confidence based on the target confidence and a third confidence, the third confidence including other confidences in the second confidence except the target confidence; determining a final confidence based on the weight value and the first confidence; and determining whether to wake up the terminal based on the final confidence.
可选地，所述根据所述目标置信度和第三置信度，获取所述第一置信度对应的权重值包括：获取所述目标置信度和所述第三置信度之间的置信度差值；根据所述置信度差值以及预设的对应关系，获取所述第一置信度对应的权重值。Optionally, obtaining the weight value corresponding to the first confidence based on the target confidence and the third confidence includes: obtaining a confidence difference between the target confidence and the third confidence; and obtaining the weight value corresponding to the first confidence based on the confidence difference and a preset corresponding relationship.
可选地，所述根据所述最终置信度确定是否唤醒所述终端包括：在所述最终置信度大于或等于预设置信度阈值的情况下，确定唤醒所述终端。Optionally, determining whether to wake up the terminal according to the final confidence level includes: determining to wake up the terminal when the final confidence level is greater than or equal to a preset confidence threshold.
可选地，所述信号处理方式包括盲源分离或噪声抑制。Optionally, the signal processing method includes blind source separation or noise suppression.
根据本公开实施例的第二方面，提供一种唤醒控制装置，所述装置包括：数据采集模块，被配置为采集多路音频数据；信号处理模块，被配置为分别对目标时间段内采集的每路第一音频数据进行信号处理，得到多个第一目标音频数据；第一置信度获取模块，被配置为分别获取多个所述第一目标音频数据的第一置信度，其中，所述置信度用于表征所述音频数据能够唤醒终端的概率；第二置信度获取模块，被配置为获取历史时间段内第二目标音频数据的第二置信度，所述历史时间段为所述目标时间段之前的预设时间段，所述第二置信度包括多个所述第二目标音频数据的置信度；唤醒模块，被配置为根据所述第一置信度和所述第二置信度确定是否唤醒所述终端。According to a second aspect of an embodiment of the present disclosure, a wake-up control device is provided, the device comprising: a data acquisition module, configured to collect multiple channels of audio data; a signal processing module, configured to perform signal processing on each channel of first audio data collected within a target time period, respectively, to obtain multiple first target audio data; a first confidence acquisition module, configured to respectively obtain first confidences of multiple first target audio data, wherein the confidences are used to characterize the probability that the audio data can wake up a terminal; a second confidence acquisition module, configured to obtain second confidences of second target audio data within a historical time period, the historical time period being a preset time period before the target time period, the second confidence including the confidences of multiple second target audio data; a wake-up module, configured to determine whether to wake up the terminal based on the first confidence and the second confidence.
可选地，所述信号处理模块包括：通道选取子模块，被配置为选取所述终端的麦克风阵列的一个麦克风作为参考通道；参考数据获取子模块，被配置为获取所述参考通道在所述目标时间段采集的参考音频数据；信号处理子模块，被配置为根据所述参考音频数据，通过多个所述信号处理方式分别对每路所述第一音频数据进行信号处理，得到多个所述第一目标音频数据，其中，不同路的第一音频数据的信号处理方式不同。Optionally, the signal processing module includes: a channel selection submodule, configured to select a microphone of the microphone array of the terminal as a reference channel; a reference data acquisition submodule, configured to obtain reference audio data collected by the reference channel during the target time period; a signal processing submodule, configured to perform signal processing on each channel of the first audio data according to the reference audio data by multiple signal processing methods, to obtain multiple first target audio data, wherein the signal processing methods for different channels of first audio data are different.
可选地，所述第一置信度获取模块包括：处理方式确定子模块，被配置为针对多个所述第一目标音频数据中的每个所述第一目标音频数据，确定该第一目标音频数据对应的信号处理方式；解码器确定子模块，被配置为根据该信号处理方式确定该第一目标音频数据对应的目标解码器，其中，不同的信号处理方式对应不同的解码器；置信度确定子模块，被配置为将该第一目标音频数据输入所述目标解码器进行解码处理，得到该第一目标音频数据的第一置信度。Optionally, the first confidence acquisition module includes: a processing method determination submodule, configured to determine, for each of the multiple first target audio data, a signal processing method corresponding to the first target audio data; a decoder determination submodule, configured to determine a target decoder corresponding to the first target audio data according to the signal processing method, wherein different signal processing methods correspond to different decoders; and a confidence determination submodule, configured to input the first target audio data into the target decoder for decoding processing to obtain a first confidence of the first target audio data.
可选地，所述解码器确定子模块，被配置为：通过预先设置的解码器关联关系，从多个所述解码器中确定该信号处理方式对应的解码器，所述解码器关联关系包括不同的信号处理方式和所述解码器的对应关系；将所述信号处理方式对应的解码器作为所述目标解码器。Optionally, the decoder determination submodule is configured to: determine the decoder corresponding to the signal processing method from multiple decoders through a pre-set decoder association relationship, the decoder association relationship including the correspondence between different signal processing methods and the decoders; and use the decoder corresponding to the signal processing method as the target decoder.
可选地，所述唤醒模块包括：唤醒处理子模块，被配置为在每获取到一个所述目标解码器输出的所述第一置信度的情况下，根据所述目标解码器输出的第一目标置信度执行以下唤醒处理方式，直至唤醒所述终端，或者，根据多个所述目标解码器输出的所述第一置信度均执行了所述唤醒处理方式；所述唤醒处理方式包括：根据所述第二置信度和所述目标解码器输出的第一置信度确定是否唤醒所述终端。Optionally, the wake-up module includes: a wake-up processing sub-module, which is configured to perform the following wake-up processing method according to the first target confidence output by the target decoder each time the first confidence output by the target decoder is obtained, until the terminal is woken up, or the wake-up processing method is performed according to the first confidence output by multiple target decoders; the wake-up processing method includes: determining whether to wake up the terminal according to the second confidence and the first confidence output by the target decoder.
可选地，所述唤醒模块包括：目标置信度确定子模块，被配置为从所述第二置信度中确定目标置信度，所述目标置信度和所述第一置信度是通过相同解码器解码得到的置信度；权重值确定子模块，被配置为根据所述目标置信度和第三置信度，获取所述第一置信度对应的权重值，所述第三置信度包括所述第二置信度中除所述目标置信度之外的其他置信度；最终置信度获取子模块，被配置为根据所述权重值和所述第一置信度，获取最终置信度；唤醒子模块，被配置为根据所述最终置信度确定是否唤醒所述终端。Optionally, the wake-up module includes: a target confidence determination submodule, configured to determine a target confidence from the second confidence, the target confidence and the first confidence are confidences obtained by decoding by the same decoder; a weight value determination submodule, configured to obtain a weight value corresponding to the first confidence based on the target confidence and a third confidence, the third confidence including other confidences in the second confidence except the target confidence; a final confidence acquisition submodule, configured to obtain a final confidence based on the weight value and the first confidence; and a wake-up submodule, configured to determine whether to wake up the terminal based on the final confidence.
可选地，所述权重值确定子模块，还被配置为：获取所述目标置信度和所述第三置信度之间的置信度差值；根据所述置信度差值以及预设的对应关系，获取所述第一置信度对应的权重值。Optionally, the weight value determination submodule is further configured to: obtain a confidence difference between the target confidence and the third confidence; and obtain a weight value corresponding to the first confidence according to the confidence difference and a preset corresponding relationship.
可选地，所述唤醒处理子模块，还被配置为：在所述最终置信度大于或等于预设置信度阈值的情况下，确定唤醒所述终端。Optionally, the wake-up processing submodule is further configured to: determine to wake up the terminal when the final confidence is greater than or equal to a preset confidence threshold.
可选地，所述信号处理方式包括盲源分离或噪声抑制。Optionally, the signal processing method includes blind source separation or noise suppression.
根据本公开实施例的第三方面，提供一种计算机可读存储介质，其上存储有计算机程序指令，该程序指令被处理器执行时实现本公开第一方面所提供的唤醒控制方法的步骤。According to a third aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the wake-up control method provided in the first aspect of the present disclosure are implemented.
根据本公开实施例的第四方面，提供一种终端，包括：存储器，其上存储有计算机程序；处理器，用于执行所述存储器中的所述计算机程序，以实现本公开第一方面所提供的唤醒控制方法的步骤。According to a fourth aspect of an embodiment of the present disclosure, a terminal is provided, comprising: a memory on which a computer program is stored; and a processor for executing the computer program in the memory to implement the steps of the wake-up control method provided in the first aspect of the present disclosure.
本公开的实施例提供的技术方案可以包括以下有益效果：通过采集多路音频数据；分别对目标时间段内采集的每路第一音频数据进行信号处理，得到多个第一目标音频数据；分别获取多个所述第一目标音频数据的第一置信度，其中，所述置信度用于表征所述音频数据能够唤醒终端的概率；获取历史时间段内第二目标音频数据的第二置信度，所述历史时间段为所述目标时间段之前的预设时间段，所述第二置信度包括多个所述第二目标音频数据的置信度；根据所述第一置信度和所述第二置信度确定是否唤醒所述终端。也就是说，本公开可以根据目标时间段的第一置信度和历史时间段的第二置信度确定是否唤醒终端，这样，可以减少终端被误唤醒或漏唤醒的概率，从而可以提高语音识别系统的准确率。The technical solution provided by the embodiments of the present disclosure may include the following beneficial effects: by collecting multiple channels of audio data; performing signal processing on each channel of first audio data collected within the target time period to obtain multiple first target audio data; obtaining the first confidence of multiple first target audio data respectively, wherein the confidence is used to characterize the probability that the audio data can wake up the terminal; obtaining the second confidence of the second target audio data within the historical time period, wherein the historical time period is a preset time period before the target time period, and the second confidence includes the confidence of multiple second target audio data; determining whether to wake up the terminal according to the first confidence and the second confidence. In other words, the present disclosure can determine whether to wake up the terminal according to the first confidence of the target time period and the second confidence of the historical time period, so that the probability of the terminal being woken up by mistake or missed can be reduced, thereby improving the accuracy of the speech recognition system.
应当理解的是，以上的一般描述和后文的细节描述仅是示例性和解释性的，并不能限制本公开。It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
此处的附图被并入说明书中并构成本说明书的一部分，示出了符合本公开的实施例，并与说明书一起用于解释本公开的原理。The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.
图1是根据一示例性实施例示出的一种唤醒控制方法的流程图；FIG1 is a flow chart showing a wake-up control method according to an exemplary embodiment;
图2是根据一示例性实施例示出的一种终端的结构示意图；FIG2 is a schematic diagram showing the structure of a terminal according to an exemplary embodiment;
图3是根据一示例性实施例示出的另一种唤醒控制方法的流程图；FIG3 is a flow chart showing another wake-up control method according to an exemplary embodiment;
图4是根据一示例性实施例示出的一种唤醒控制装置的结构示意图；FIG4 is a schematic structural diagram of a wake-up control device according to an exemplary embodiment;
图5是根据一示例性实施例示出的一种终端的框图。Fig. 5 is a block diagram of a terminal according to an exemplary embodiment.
具体实施方式Detailed ways
这里将详细地对示例性实施例进行说明，其示例表示在附图中。下面的描述涉及附图时，除非另有表示，不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本公开相一致的所有实施方式。相反，它们仅是与如所附权利要求书中所详述的、本公开的一些方面相一致的装置和方法的例子。Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.
首先，对本公开的应用场景进行介绍。本公开可以应用于具有语音识别功能的终端，在现实环境中，由于环境噪声对语音识别系统的影响，终端被误唤醒或漏唤醒的概率较高。考虑到单一麦克风无法有效地处理噪声，尤其是频率响应随时间而改变的噪声，例如音乐，相关技术中，为了满足不同的噪声场景，例如背景人声、洗衣机、电视机等，可以在终端中采用麦克风阵列，并通过多个解码器进行预测，确定是否唤醒该终端。First, the application scenarios of the present disclosure are introduced. The present disclosure can be applied to terminals with voice recognition functions. In real environments, due to the influence of environmental noise on the voice recognition system, the probability of the terminal being woken up by mistake or missed is high. Considering that a single microphone cannot effectively process noise, especially noise whose frequency response changes over time, such as music, in related technologies, in order to meet different noise scenarios, such as background human voices, washing machines, televisions, etc., a microphone array can be used in the terminal, and multiple decoders can be used for prediction to determine whether to wake up the terminal.
但是，通过多个解码器进行预测时，只要任一解码器输出的预测结果是唤醒终端，则确定对该终端进行唤醒。在这种情况下，如果该解码器预测的准确率较低，则会导致该终端被误唤醒，从而使得该终端的语音识别系统的准确率较低，影响用户体验。However, when multiple decoders are used for prediction, as long as the prediction result output by any decoder is to wake up the terminal, the terminal is determined to be woken up. In this case, if the accuracy of the decoder prediction is low, the terminal will be woken up by mistake, thereby making the accuracy of the voice recognition system of the terminal low, affecting the user experience.
为了解决上述问题，本公开提供一种唤醒控制方法、装置、存储介质及终端，通过分别对目标时间段内采集的每路第一音频数据进行信号处理，得到多个第一目标音频数据，并分别获取该多个第一目标音频数据的第一置信度，之后，可以获取历史时间段内第二目标音频数据的第二置信度，并根据该第一置信度和该第二置信度确定是否唤醒该终端。也就是说，本公开可以根据目标时间段的第一置信度和历史时间段的第二置信度确定是否唤醒终端，这样，可以减少终端被误唤醒或漏唤醒的概率，从而可以提高语音识别系统的准确率。In order to solve the above problems, the present disclosure provides a wake-up control method, device, storage medium and terminal, which respectively processes the signal of each channel of first audio data collected in the target time period to obtain multiple first target audio data, and respectively obtains the first confidence of the multiple first target audio data, and then obtains the second confidence of the second target audio data in the historical time period, and determines whether to wake up the terminal according to the first confidence and the second confidence. In other words, the present disclosure can determine whether to wake up the terminal according to the first confidence of the target time period and the second confidence of the historical time period, so that the probability of the terminal being woken up by mistake or missed can be reduced, thereby improving the accuracy of the speech recognition system.
下面结合具体实施例对本公开进行说明。The present disclosure is described below in conjunction with specific embodiments.
图1是根据一示例性实施例示出的一种唤醒控制方法的流程图，如图1所示，该方法包括：FIG. 1 is a flow chart of a wake-up control method according to an exemplary embodiment. As shown in FIG. 1 , the method includes:
S101、采集多路音频数据。S101, collecting multiple channels of audio data.
需要说明的是，本唤醒控制方法应用于语音交互功能的终端设备，比如，终端设备安装有具有语音交互功能的应用，如语音助手应用，该语音助手应用用于识别用户的语音信息。本公开实施例可以应用于各种终端设备中，该终端设备包括但不限于固定设备和移动设备，例如，所述固定设备包括但不限于：个人电脑(Personal Computer，PC)、电视、空调、壁挂炉等；所述移动设备包括但不限于：手机、平板电脑、可穿戴式设备、音箱、闹钟等，本公开对此不作限定。图2是根据一示例性实施例示出的一种终端的结构示意图，如图2所示，该终端可以包括麦克风阵列、信号处理模块、解码器以及唤醒模块，该麦克风阵列可以包括多个麦克风，该解码器也可以包括多个。该终端可以通过该麦克风阵列实时采集多路音频数据，并将该多路音频数据发送至该信号处理模块，该信号处理模块可以根据对该多路音频数据进行处理，得到处理后的多个目标音频数据，之后，可以通过多个解码器对该多个目标音频数据进行解码处理，得到多个置信度，最后根据该置信度确定是否唤醒该终端。It should be noted that the wake-up control method is applied to a terminal device with a voice interaction function. For example, the terminal device is installed with an application with a voice interaction function, such as a voice assistant application, which is used to recognize the user's voice information. The embodiment of the present disclosure can be applied to various terminal devices, including but not limited to fixed devices and mobile devices. For example, the fixed device includes but is not limited to: a personal computer (PC), a television, an air conditioner, a wall-mounted boiler, etc.; the mobile device includes but is not limited to: a mobile phone, a tablet computer, a wearable device, a speaker, an alarm clock, etc., and the present disclosure does not limit this. Figure 2 is a structural schematic diagram of a terminal according to an exemplary embodiment. As shown in Figure 2, the terminal may include a microphone array, a signal processing module, a decoder, and a wake-up module. The microphone array may include multiple microphones, and the decoder may also include multiple. The terminal can collect multiple audio data in real time through the microphone array, and send the multiple audio data to the signal processing module. The signal processing module can process the multiple audio data to obtain multiple target audio data after processing. After that, the multiple target audio data can be decoded by multiple decoders to obtain multiple confidences, and finally determine whether to wake up the terminal according to the confidence.
在本步骤中，在该终端开机启动后，该终端的采集模块可以通过麦克风阵列中的多个麦克风采集多路音频数据，每个麦克风对应一路音频数据。In this step, after the terminal is powered on, the acquisition module of the terminal can collect multiple channels of audio data through multiple microphones in the microphone array, and each microphone corresponds to one channel of audio data.
S102、分别对目标时间段内采集的每路第一音频数据进行信号处理，得到多个第一目标音频数据。S102: Perform signal processing on each channel of first audio data collected within a target time period to obtain a plurality of first target audio data.
在本步骤中，可以根据终端所使用的环境，预先设置对应的信号处理方式，示例地，针对经常在吵杂环境使用的终端，例如手机，可以设置较多的信号处理方式，针对在比较安静的环境使用的终端，例如空调，可以设置较少的信号处理方式。In this step, the corresponding signal processing method can be pre-set according to the environment in which the terminal is used. For example, for terminals that are often used in noisy environments, such as mobile phones, more signal processing methods can be set, and for terminals used in relatively quiet environments, such as air conditioners, fewer signal processing methods can be set.
在获取到目标时间段内采集的多路第一音频数据后，可以通过该终端预先设置的信号处理方式对该第一音频数据进行信号处理，得到多个第一目标音频数据。After acquiring multiple channels of first audio data collected within the target time period, the first audio data may be processed by a signal processing method preset by the terminal to obtain multiple first target audio data.
S103、分别获取多个第一目标音频数据的第一置信度。S103: Obtain first confidences of multiple first target audio data respectively.
其中，该置信度用于表征该音频数据能够唤醒终端的概率，该置信度的范围可以是0～100。The confidence level is used to represent the probability that the audio data can wake up the terminal, and the confidence level may range from 0 to 100.
在本步骤中，在获取该多个第一目标音频数据后，可以针对每个第一目标音频数据，通过解码器对该第一目标音频数据进行解码处理，得到该第一目标音频数据的第一置信度，最终得到多个第一置信度。In this step, after obtaining the multiple first target audio data, each first target audio data can be decoded by a decoder to obtain the first confidence of the first target audio data, and finally obtain multiple first confidences.
S104、获取历史时间段内第二目标音频数据的第二置信度。S104: Obtain a second confidence level of the second target audio data within a historical time period.
其中，该历史时间段为该目标时间段之前的预设时间段，该历史时间段可以是与该目标时间段属于相同场景的时间段，例如历史时间段和目标时间段都属于语音唤醒场景下采集音频数据的时间段。另外，也可以根据终端的类型设置该预设时间段的时长，也可以根据测试经验值设置该预设时间段的时长，本公开对此不作限定。Among them, the historical time period is a preset time period before the target time period, and the historical time period may be a time period belonging to the same scenario as the target time period, for example, the historical time period and the target time period both belong to the time period for collecting audio data in the voice wake-up scenario. In addition, the length of the preset time period may also be set according to the type of terminal, or according to the test experience value, which is not limited in this disclosure.
该第二置信度可以包括多个第二目标音频数据的置信度，该多个第二目标音频数据为对该第二音频数据进行信号处理后的音频数据。该第二置信度的获取方式可以参考该第一置信度的获取方式，此处不再赘述了。另外，该第二置信度可以存储在该终端中，例如，若该预设时间段为1分钟，则可以存储该目标时间段之前的1分钟的第二目标音频数据的第二置信度。示例地，本公开可以采用队列的方式存储该第二置信度，不同的第二置信度可以对应不同的队列，例如，若该第二置信度包括10个，则可以使用10个队列存储该10个第二置信度。针对每个队列，该队列中队头的第二置信度为最早获取的第二目标音频数据的第二置信度，队尾的第二置信度为最晚获取的第二目标音频数据的第二置信度，在需要存储新的第二置信度时，可以将该队列中队头的第二置信度删除，将该新的第二置信度存储在该队列的队尾。The second confidence may include the confidence of multiple second target audio data, and the multiple second target audio data are the audio data after the second audio data is subjected to signal processing. The acquisition method of the second confidence may refer to the acquisition method of the first confidence, which will not be repeated here. In addition, the second confidence may be stored in the terminal. For example, if the preset time period is 1 minute, the second confidence of the second target audio data 1 minute before the target time period may be stored. For example, the present disclosure may store the second confidence in the form of a queue, and different second confidences may correspond to different queues. For example, if the second confidence includes 10, 10 queues may be used to store the 10 second confidences. For each queue, the second confidence at the head of the queue is the second confidence of the second target audio data obtained earliest, and the second confidence at the tail of the queue is the second confidence of the second target audio data obtained latest. When a new second confidence needs to be stored, the second confidence at the head of the queue may be deleted, and the new second confidence may be stored at the tail of the queue.
在本步骤中，在获取该多个第一目标音频数据的第一置信度后，可以获取存储的多个第二置信度。In this step, after obtaining the first confidences of the plurality of first target audio data, the plurality of stored second confidences may be obtained.
S105、根据该第一置信度和该第二置信度确定是否唤醒该终端。S105: Determine whether to wake up the terminal according to the first confidence level and the second confidence level.
在本步骤中，在得到该多个第一目标音频数据的多个第一置信度后，可以根据任意一个第一置信度和该第二置信度确定是否唤醒该终端。考虑到一段时间内的环境噪声的变化较小，为了避免该第一置信度不够准确导致该终端被误唤醒或漏唤醒，可以根据该第二置信度调整该第一置信度。示例地，若得到的该第一置信度比较高，而在该第一置信度对应的第一目标音频数据之前的历史时间段获取的第二目标音频数据的第二置信度比较低，则该第一置信度可能会存在较大误差，这种情况下，可以参考该第二置信度降低该第一置信度，从而可以得到更加准确的第一置信度。In this step, after obtaining multiple first confidences of the multiple first target audio data, it can be determined whether to wake up the terminal based on any one of the first confidences and the second confidence. Considering that the change in ambient noise over a period of time is small, in order to avoid the terminal being mistakenly awakened or missed awakening due to the first confidence being inaccurate, the first confidence can be adjusted according to the second confidence. For example, if the first confidence obtained is relatively high, and the second confidence of the second target audio data obtained in the historical time period before the first target audio data corresponding to the first confidence is relatively low, the first confidence may have a large error. In this case, the first confidence can be lowered with reference to the second confidence, so that a more accurate first confidence can be obtained.
进一步地，在得到调整后的第一置信度后，可以根据该调整后的第一置信度确定是否唤醒该终端，示例地，在该调整后的第一置信度较高的情况下，可以确定唤醒该终端，在该调整后的第一置信度较低的情况下，可以确定不唤醒该终端。Further, after obtaining the adjusted first confidence, it can be determined whether to wake up the terminal based on the adjusted first confidence. For example, when the adjusted first confidence is high, it can be determined to wake up the terminal. When the adjusted first confidence is low, it can be determined not to wake up the terminal.
采用上述方法，通过分别对目标时间段内采集的每路第一音频数据进行信号处理，得到多个第一目标音频数据，并分别获取该多个第一目标音频数据的第一置信度，之后，可以获取历史时间段内第二目标音频数据的第二置信度，并根据该第一置信度和该第二置信度确定是否唤醒该终端。也就是说，本公开可以根据目标时间段的第一置信度和历史时间段的第二置信度确定是否唤醒终端，这样，可以减少终端被误唤醒或漏唤醒的概率，从而可以提高语音识别系统的准确率。By adopting the above method, by performing signal processing on each channel of first audio data collected in the target time period, multiple first target audio data are obtained, and the first confidence of the multiple first target audio data is obtained respectively. After that, the second confidence of the second target audio data in the historical time period can be obtained, and whether to wake up the terminal is determined based on the first confidence and the second confidence. In other words, the present disclosure can determine whether to wake up the terminal based on the first confidence of the target time period and the second confidence of the historical time period. In this way, the probability of the terminal being woken up by mistake or missed can be reduced, thereby improving the accuracy of the speech recognition system.
图3是根据一示例性实施例示出的另一种唤醒控制方法的流程图，如图3所示，该方法包括：FIG. 3 is a flow chart of another wake-up control method according to an exemplary embodiment. As shown in FIG. 3 , the method includes:
S301、采集多路音频数据。S301, collecting multiple channels of audio data.
S302、选取该终端的麦克风阵列的一个麦克风作为参考通道。S302: Select a microphone of the microphone array of the terminal as a reference channel.
需要说明的是，在该终端通过采集模块采集该多路音频数据的同时，该终端可能也在输出音频数据，例如，该终端正在播放音乐、播放视频或者播放铃音等，在这种情况下，该终端采集的多路音频数据中也会包括该终端输出的音频数据。而该终端在进行语音识别时，需要从采集的音频数据中提取用户输入的语音，因此，如图2所示，可以将该终端的麦克风阵列中的一个麦克风作为参考通道，通过该参考通道获取该终端输出的音频数据。It should be noted that while the terminal is collecting the multi-channel audio data through the collection module, the terminal may also be outputting audio data. For example, the terminal is playing music, playing video or playing ringtones, etc. In this case, the multi-channel audio data collected by the terminal will also include the audio data output by the terminal. When the terminal performs speech recognition, it is necessary to extract the user input speech from the collected audio data. Therefore, as shown in FIG2 , one microphone in the microphone array of the terminal can be used as a reference channel, and the audio data output by the terminal can be obtained through the reference channel.
S303、获取该参考通道在该目标时间段采集的参考音频数据。S303: Acquire reference audio data collected by the reference channel in the target time period.
在本步骤中，该终端在采集该多路音频数据的同时，也可以通过该参考通道实时获取该终端输出的参考音频数据，这样，该终端在获取目标时间段的第一音频数据后，也同步获取到了该目标时间段的参考音频数据。In this step, while collecting the multi-channel audio data, the terminal can also obtain the reference audio data output by the terminal in real time through the reference channel. In this way, after obtaining the first audio data of the target time period, the terminal also synchronously obtains the reference audio data of the target time period.
S304、根据该参考音频数据，通过多个信号处理方式分别对每路该第一音频数据进行信号处理，得到多个该第一目标音频数据。S304: Perform signal processing on each channel of the first audio data respectively through multiple signal processing methods according to the reference audio data to obtain multiple first target audio data.
其中，不同路的第一音频数据的信号处理方式不同，该信号处理方式可以包括盲源分离或噪声抑制。The signal processing methods of the first audio data of different channels are different, and the signal processing method may include blind source separation or noise suppression.
在本步骤中，在获取到该目标时间段内采集的多路第一音频数据后，可以根据该参考音频数据，通过该终端预先设置的该多个信号处理方式对该多路第一音频数据进行信号处理，滤除该第一音频数据中的噪声，得到该第一音频数据中的语音，即第一目标音频数据。示例地，若该第一音频数据包括两路，该信号处理方式包括盲源分离和噪声抑制两种方式，则可以根据该参考音频数据，分别通过该盲源分离方式和该噪声抑制方式对该两路第一音频数据进行信号处理，得到两个第一目标音频数据。由于同步获取了该终端在该目标时段输出的该参考音频数据，因此，在对该第一音频数据进行信号处理时，可以根据该参考音频数据滤除该第一音频数据中的部分噪声，这样，可以简化信号处理的复杂度，从而可以减少该终端的唤醒延迟，提高用户体验。In this step, after acquiring the multiple channels of first audio data collected during the target time period, the multiple channels of first audio data can be processed according to the reference audio data by the multiple signal processing methods pre-set by the terminal, and the noise in the first audio data can be filtered out to obtain the voice in the first audio data, that is, the first target audio data. For example, if the first audio data includes two channels, and the signal processing method includes two methods of blind source separation and noise suppression, the two channels of first audio data can be processed by the blind source separation method and the noise suppression method respectively according to the reference audio data to obtain two first target audio data. Since the reference audio data output by the terminal in the target time period is synchronously acquired, when the first audio data is processed, part of the noise in the first audio data can be filtered out according to the reference audio data. In this way, the complexity of signal processing can be simplified, thereby reducing the wake-up delay of the terminal and improving the user experience.
S305、针对多个第一目标音频数据中的每个第一目标音频数据，确定该第一目标音频数据对应的信号处理方式。S305: For each first target audio data among the plurality of first target audio data, determine a signal processing method corresponding to the first target audio data.
其中，不同路的第一音频数据对应不同的信号处理方式。The first audio data of different channels correspond to different signal processing methods.
在本步骤中，在采集到每路第一音频数据后，都会通过对应的信号处理方式对该路第一音频数据进行信号处理，得到该路第一音频数据对应的第一目标音频数据，因此，针对每个第一目标音频数据，都会对应一个信号处理方式。在得到该多个第一目标音频数据后，可以针对每个第一目标音频数据，确定该第一目标音频数据对应的信号处理方式。In this step, after collecting each channel of first audio data, the first audio data is processed by a corresponding signal processing method to obtain the first target audio data corresponding to the first audio data. Therefore, a signal processing method corresponds to each first target audio data. After obtaining the plurality of first target audio data, the signal processing method corresponding to the first target audio data can be determined for each first target audio data.
S306、根据该信号处理方式确定该第一目标音频数据对应的目标解码器。S306: Determine a target decoder corresponding to the first target audio data according to the signal processing method.
其中，不同的信号处理方式对应不同的解码器，不同的解码器的参数可以不同，示例地，该参数可以根据信号处理方式的类型确定，针对不同类型的信号处理方式，可以设置不同的参数，本公开对此不作限定。Among them, different signal processing methods correspond to different decoders, and the parameters of different decoders may be different. For example, the parameters can be determined according to the type of signal processing method. Different parameters can be set for different types of signal processing methods, and the present disclosure does not limit this.
在本步骤中，在确定该第一目标音频数据对应的信号处理方式后，可以通过预先设置的解码器关联关系，从多个解码器中确定该信号处理方式对应的解码器，该解码器关联关系可以包括不同的信号处理方式和解码器的对应关系，将该信号处理方式对应的解码器作为该目标解码器。In this step, after determining the signal processing method corresponding to the first target audio data, the decoder corresponding to the signal processing method can be determined from multiple decoders through a pre-set decoder association relationship. The decoder association relationship can include a correspondence between different signal processing methods and decoders, and the decoder corresponding to the signal processing method is used as the target decoder.
S307、将该第一目标音频数据输入该目标解码器进行解码处理，并输出该第一目标音频数据的第一置信度。S307: Input the first target audio data into the target decoder for decoding processing, and output a first confidence level of the first target audio data.
在本步骤中，在获取每个第一目标音频数据对应的目标解码器后，针对该第一目标音频数据，可以将该第一目标音频数据输入该目标解码器中，通过该目标解码器对该第一目标音频数据进行解码处理，得到该第一目标音频数据的第一置信度。In this step, after obtaining the target decoder corresponding to each first target audio data, the first target audio data can be input into the target decoder, and the first target audio data can be decoded by the target decoder to obtain the first confidence of the first target audio data.
S308、获取历史时间段内第二目标音频数据的第二置信度。S308: Obtain a second confidence level of the second target audio data within the historical time period.
其中，该历史时间段为该目标时间段之前的预设时间段，该历史时间段可以是与该目标时间段属于相同场景的时间段，可以根据终端的类型设置该预设时间段的时长，也可以根据测试经验值设置该预设时间段的时长，本公开对此不作限定。Among them, the historical time period is a preset time period before the target time period. The historical time period may be a time period belonging to the same scenario as the target time period. The length of the preset time period may be set according to the type of terminal or according to test experience values. The present disclosure does not limit this.
该第二置信度可以包括多个第二目标音频数据的置信度，该多个第二目标音频数据为对该第二音频数据进行信号处理后的音频数据。该第二置信度的获取方式可以参考该第一置信度的获取方式，此处不再赘述了，另外，该第二置信度可以存储在该终端中，例如，若该预设时间段为1分钟，则可以存储该目标时间段之前的1分钟的第二目标音频数据的第二置信度。示例地，本公开可以采用队列的方式存储该第二置信度，不同的第二置信度可以对应不同的队列，例如，若该第二置信度包括10个，则可以使用10个队列存储该10个第二置信度。针对每个队列，该队列中队头的第二置信度为最早获取的第二目标音频数据的第二置信度，队尾的第二置信度为最晚获取的第二目标音频数据的第二置信度，在需要存储新的第二置信度时，可以将该队列中队头的第二置信度删除，将该新的第二置信度存储在该队列的队尾。The second confidence may include the confidence of multiple second target audio data, and the multiple second target audio data are the audio data after the second audio data is subjected to signal processing. The acquisition method of the second confidence may refer to the acquisition method of the first confidence, which will not be repeated here. In addition, the second confidence may be stored in the terminal. For example, if the preset time period is 1 minute, the second confidence of the second target audio data 1 minute before the target time period may be stored. For example, the present disclosure may store the second confidence in a queue manner, and different second confidences may correspond to different queues. For example, if the second confidence includes 10, 10 queues may be used to store the 10 second confidences. For each queue, the second confidence of the head of the queue is the second confidence of the second target audio data obtained earliest, and the second confidence of the tail is the second confidence of the second target audio data obtained latest. When a new second confidence needs to be stored, the second confidence of the head of the queue may be deleted, and the new second confidence may be stored at the tail of the queue.
S309、从该第二置信度中确定目标置信度。S309: Determine a target confidence level from the second confidence level.
其中，该目标置信度和该第一置信度是通过相同解码器解码得到的置信度。The target confidence and the first confidence are confidences obtained by decoding through the same decoder.
在本步骤中，在确定该目标置信度之前，先确定该第一置信度对应的目标解码器，根据该第一置信度对应的目标解码器，从该第二置信度中，确定与该第一置信度对应的目标解码器相同的目标置信度。In this step, before determining the target confidence, the target decoder corresponding to the first confidence is first determined, and based on the target decoder corresponding to the first confidence, a target confidence that is the same as the target decoder corresponding to the first confidence is determined from the second confidence.
S310、根据该目标置信度和第三置信度，获取该第一置信度对应的权重值。S310: Obtain a weight value corresponding to the first confidence level according to the target confidence level and the third confidence level.
其中，该第三置信度包括该第二置信度中除该目标置信度之外的其他置信度。The third confidence level includes other confidence levels in the second confidence level except the target confidence level.
在本步骤中，在获取该目标置信度后，可以根据该目标置信度，确定该第三置信度，之后，可以获取该目标置信度和该第三置信度之间的置信度差值，根据该置信度差值以及预设的对应关系，获取该第一置信度对应的权重值。其中，若该第三置信度包括一个置信度，则可以直接计算该目标置信度和该第三置信度的置信度差值；若该第三置信度包括多个置信度，则可以先获取该多个第三置信度的平均置信度，再计算该目标置信度和该平均置信度的置信度差值。In this step, after obtaining the target confidence, the third confidence can be determined according to the target confidence, and then the confidence difference between the target confidence and the third confidence can be obtained, and the weight value corresponding to the first confidence can be obtained according to the confidence difference and the preset corresponding relationship. If the third confidence includes one confidence, the confidence difference between the target confidence and the third confidence can be directly calculated; if the third confidence includes multiple confidences, the average confidence of the multiple third confidences can be obtained first, and then the confidence difference between the target confidence and the average confidence can be calculated.
在一种可能的实现方式中，该预设的对应关系可以是预先设置的权重值关系，该权重值关系包括该置信度差值和该权重值的对应关系，在获取该目标置信度和该第三置信度之间的置信度差值后，可以通过该权重值关系，确定该置信度差值对应的权重值。其中，该权重值关系可以根据经验进行设置，示例地，在该置信度差值为0.15时，该权重值可以是1.1，在该置信度差值为-0.2时，该权重值可以是0.8。In a possible implementation, the preset correspondence relationship may be a preset weight value relationship, the weight value relationship including the correspondence relationship between the confidence difference and the weight value, and after obtaining the confidence difference between the target confidence and the third confidence, the weight value corresponding to the confidence difference may be determined by the weight value relationship. The weight value relationship may be set based on experience, for example, when the confidence difference is 0.15, the weight value may be 1.1, and when the confidence difference is -0.2, the weight value may be 0.8.
在另一种可能的实现方式中，任何一路解码器权重值的预设的对应关系可以是以下计算公式：In another possible implementation, the preset corresponding relationship of any decoder weight value may be the following calculation formula:
ratio＝1+(Asmooth-Bsmooth)/a (1)ratio＝1+(Asmooth -Bsmooth )/a (1)
其中，ratio为该权重值，Asmooth为该目标置信度，Bsmooth为该第三置信度，a为预设常数。Among them, ratio is the weight value, Asmooth is the target confidence, Bsmooth is the third confidence, and a is a preset constant.
在获取该目标置信度和该第三置信度之间的置信度差值后，可以根据该置信度差值和预设常数，通过公式(1)计算得到该第一置信度对应的权重值。After obtaining the confidence difference between the target confidence and the third confidence, the weight value corresponding to the first confidence can be calculated according to the confidence difference and a preset constant by formula (1).
S311、根据该权重值和该第一置信度，获取最终置信度。S311. Obtain a final confidence level according to the weight value and the first confidence level.
在本步骤中的一些实施例中，在获取该权重值后，可以为该第一置信度乘以该权重值，得到该最终置信度。在另一些实施例中，该最终置信度也可以是该第一置信度与该权重值的和值、差值或除值。本公开对于如何根据权重值和第一置信度获取最终置信度的方式不作限定，可以根据不同的需要进行设定。In some embodiments of this step, after obtaining the weight value, the first confidence level may be multiplied by the weight value to obtain the final confidence level. In other embodiments, the final confidence level may also be the sum, difference, or division of the first confidence level and the weight value. The present disclosure does not limit how to obtain the final confidence level based on the weight value and the first confidence level, and it may be set according to different needs.
在该目标置信度大于该第三置信度时，得到的该最终置信度也会大于该目标解码器输出的第一置信度，在该目标置信度小于该第三置信度时，得到的该最终置信度也会小于该目标解码器输出的第一置信度。这样，可以通过历史时间段的第二置信度对该第一置信度进行修正，从而可以得到更加准确的置信度，提高该终端的语音识别系统的准确率。When the target confidence is greater than the third confidence, the final confidence obtained will also be greater than the first confidence output by the target decoder, and when the target confidence is less than the third confidence, the final confidence obtained will also be less than the first confidence output by the target decoder. In this way, the first confidence can be corrected by the second confidence of the historical time period, so that a more accurate confidence can be obtained, thereby improving the accuracy of the speech recognition system of the terminal.
S312、根据该最终置信度确定是否唤醒该终端。S312: Determine whether to wake up the terminal according to the final confidence level.
在本步骤中，在获取该最终置信度后，可以先获取预设置信度阈值，对比该最终置信度和该置信度阈值，在该最终置信度大于或等于该预设置信度阈值的情况下，确定唤醒该终端。其中，该预设置信度阈值可以根据该终端的类型确定，示例地，针对唤醒率要求较高但误报率要求较低的终端，可以设置较低的预设置信度阈值，例如0.7，针对唤醒率要求较低但误报率要求较高的终端，可以设置较高的预设置信度阈值，例如0.9，也可以根据终端对唤醒率和误报率的需求，经过试验测试确定，本公开对该预设置信度阈值的设置方式不作限定。In this step, after obtaining the final confidence, a preset confidence threshold can be obtained first, and the final confidence and the confidence threshold can be compared. When the final confidence is greater than or equal to the preset confidence threshold, it is determined to wake up the terminal. The preset confidence threshold can be determined according to the type of the terminal. For example, for a terminal with a high wake-up rate requirement but a low false alarm rate requirement, a lower preset confidence threshold, such as 0.7, can be set. For a terminal with a low wake-up rate requirement but a high false alarm rate requirement, a higher preset confidence threshold, such as 0.9, can be set. It can also be determined based on the terminal's requirements for the wake-up rate and the false alarm rate after experimental testing. The present disclosure does not limit the setting method of the preset confidence threshold.
需要说明的是，上述步骤S308～步骤S312中的第一置信度可以是多个第一置信度中的任意一个，但是，考虑到在获取目标时间段内的第一音频数据后，可以通过多个信号处理方式对该第一音频数据进行信号处理得到多个第一目标音频数据，之后，可以将该多个第一目标音频数据输入多个目标解码器，以得到该多个第一目标音频数据的第一置信度。由于多个信号处理方式对该第一音频数据进行信号处理所花费的时间不同，导致得到每个第一目标音频数据的时间不同，因此，每个第一目标音频数据输入目标解码器的时间也不同，并且每个目标解码器对该第一目标音频数据进行解码处理所花费的时间也不同，这样，最终每个目标解码器输出该第一目标音频数据的第一置信度的时间也不同。It should be noted that the first confidence in the above steps S308 to S312 can be any one of a plurality of first confidences. However, considering that after obtaining the first audio data within the target time period, the first audio data can be processed by a plurality of signal processing methods to obtain a plurality of first target audio data. After that, the plurality of first target audio data can be input into a plurality of target decoders to obtain the first confidence of the plurality of first target audio data. Since the time taken by the plurality of signal processing methods to process the first audio data is different, the time taken to obtain each first target audio data is different. Therefore, the time taken for each first target audio data to be input into the target decoder is also different, and the time taken for each target decoder to decode the first target audio data is also different. In this way, the time taken for each target decoder to output the first confidence of the first target audio data is also different.
基于上述原因，若任意选择的第一置信度输出的较慢，则会导致唤醒延迟时间比较长，影响用户体验。因此，为了避免唤醒延迟时间太长导致用户体验太差，可以在每获取到一个目标解码器输出的第一置信度的情况下，根据该目标解码器输出的第一目标置信度执行以下唤醒处理方式，直至唤醒所述终端，或者，根据多个所述目标解码器输出的所述第一置信度均执行了该唤醒处理方式。Based on the above reasons, if the output of the arbitrarily selected first confidence is slow, it will lead to a long wake-up delay time, affecting the user experience. Therefore, in order to avoid the wake-up delay time being too long and causing a poor user experience, the following wake-up processing method can be performed according to the first target confidence output by the target decoder each time the first confidence output by the target decoder is obtained, until the terminal is awakened, or the wake-up processing method is performed according to the first confidence output by multiple target decoders.
其中，该唤醒处理方式包括：根据该第二置信度和该目标解码器输出的第一置信度确定是否唤醒该终端。The wake-up processing method includes: determining whether to wake up the terminal according to the second confidence and the first confidence output by the target decoder.
示例地，若该终端包括目标解码器A、目标解码器B以及目标解码器C，则在目标解码器B先输出第一置信度的情况下，可以先根据该目标解码器B输出的第一置信度执行该唤醒处理方式。若执行该唤醒处理方式后确定唤醒该终端，则停止执行该唤醒处理方式，可以直接唤醒该终端；若执行该唤醒处理方式后确定不唤醒该终端，则可以获取下一个目标解码器输出的第一置信度，若下一个输出第一置信度的是目标解码器A，则可以根据该目标解码器A输出的第一置信度继续执行该唤醒处理方式，若根据该唤醒处理方式确定唤醒该终端，则停止执行该唤醒处理方式，可以直接唤醒该终端；若执行该唤醒处理方式后确定不唤醒该终端，则可以获取目标解码器C输出的第一置信度，根据该目标解码器C输出的第一置信度继续执行该唤醒处理方式，确定是否唤醒该终端。这样，无需等待特定的目标解码器输出第一置信度，可以根据先输出的第一置信度确定是否唤醒该终端，从而可以缩短唤醒延迟的时间，提高用户体验。For example, if the terminal includes a target decoder A, a target decoder B, and a target decoder C, then when the target decoder B outputs the first confidence first, the wake-up processing method can be first executed according to the first confidence output by the target decoder B. If the wake-up processing method is executed to determine that the terminal is to be awakened, the wake-up processing method is stopped and the terminal can be directly awakened; if the wake-up processing method is executed to determine that the terminal is not to be awakened, the first confidence output by the next target decoder can be obtained, and if the next output of the first confidence is the target decoder A, the wake-up processing method can be continued to be executed according to the first confidence output by the target decoder A, and if the terminal is determined to be awakened according to the wake-up processing method, the wake-up processing method is stopped and the terminal can be directly awakened; if the wake-up processing method is executed to determine that the terminal is not to be awakened, the first confidence output by the target decoder C can be obtained, and the wake-up processing method is continued to be executed according to the first confidence output by the target decoder C to determine whether to wake up the terminal. In this way, there is no need to wait for a specific target decoder to output the first confidence, and whether to wake up the terminal can be determined according to the first confidence output first, thereby shortening the wake-up delay time and improving the user experience.
采用上述方法，可以根据该目标解码器输出的第一置信度，从该第二置信度中确定目标置信度，根据该目标置信度和第三置信度获取该第一置信度对应的权重值，根据该权重值和该第一置信度确定最终置信度，并根据该目标置信度确定是否唤醒该终端，这样，可以根据该目标置信度和该第三置信度调整该第一置信度，从而可以得到更加准确的最终置信度，使得该终端的语音识别系统的准确率更高；另外，本公开在每获取一个目标解码器输出的第一置信度后，都会执行唤醒处理方式，从而可以提高该终端的唤醒效率，提高用户体验。By adopting the above method, the target confidence can be determined from the second confidence according to the first confidence output by the target decoder, the weight value corresponding to the first confidence can be obtained according to the target confidence and the third confidence, the final confidence can be determined according to the weight value and the first confidence, and whether to wake up the terminal can be determined according to the target confidence. In this way, the first confidence can be adjusted according to the target confidence and the third confidence, so that a more accurate final confidence can be obtained, so that the accuracy of the speech recognition system of the terminal is higher; in addition, the present disclosure executes the wake-up processing method after each first confidence output by a target decoder is obtained, thereby improving the wake-up efficiency of the terminal and improving the user experience.
图4是根据一示例性实施例示出的一种唤醒控制装置的结构示意图，如图4所示，该装置包括：FIG. 4 is a schematic diagram of a structure of a wake-up control device according to an exemplary embodiment. As shown in FIG. 4 , the device includes:
数据采集模块401，被配置为采集多路音频数据；The data acquisition module 401 is configured to acquire multiple channels of audio data;
信号处理模块402，被配置为分别对目标时间段内采集的每路第一音频数据进行信号处理，得到多个第一目标音频数据；The signal processing module 402 is configured to perform signal processing on each channel of first audio data collected within a target time period to obtain a plurality of first target audio data;
第一置信度获取模块403，被配置为分别获取多个第一目标音频数据的第一置信度，其中，该置信度用于表征该音频数据能够唤醒终端的概率；A first confidence acquisition module 403 is configured to respectively acquire first confidences of a plurality of first target audio data, wherein the confidences are used to characterize the probability that the audio data can wake up the terminal;
第二置信度获取模块404，被配置为获取历史时间段内第二目标音频数据的第二置信度，该历史时间段为该目标时间段之前的预设时间段，该第二置信度包括多个第二目标音频数据的置信度；A second confidence acquisition module 404 is configured to acquire a second confidence of the second target audio data in a historical time period, where the historical time period is a preset time period before the target time period, and the second confidence includes confidences of a plurality of second target audio data;
唤醒模块405，被配置为根据该第一置信度和该第二置信度确定是否唤醒该终端。The wake-up module 405 is configured to determine whether to wake up the terminal according to the first confidence level and the second confidence level.
可选地，该信号处理模块402包括：Optionally, the signal processing module 402 includes:
通道选取子模块，被配置为选取该终端的麦克风阵列的一个麦克风作为参考通道；A channel selection submodule is configured to select a microphone of the microphone array of the terminal as a reference channel;
参考数据获取子模块，被配置为获取该参考通道在该目标时间段采集的参考音频数据；A reference data acquisition submodule is configured to acquire reference audio data collected by the reference channel during the target time period;
信号处理子模块，被配置为根据该参考音频数据，通过多个信号处理方式分别对每路第一音频数据进行信号处理，得到多个第一目标音频数据，其中，不同路的第一音频数据的信号处理方式不同。The signal processing submodule is configured to perform signal processing on each channel of first audio data respectively through multiple signal processing methods according to the reference audio data to obtain multiple first target audio data, wherein the signal processing methods of different channels of first audio data are different.
可选地，该第一置信度获取模块403包括：Optionally, the first confidence acquisition module 403 includes:
处理方式确定子模块，被配置为针对多个第一目标音频数据中的每个第一目标音频数据，确定该第一目标音频数据对应的信号处理方式；A processing mode determination submodule is configured to determine, for each first target audio data among the plurality of first target audio data, a signal processing mode corresponding to the first target audio data;
解码器确定子模块，被配置为根据该信号处理方式确定该第一目标音频数据对应的目标解码器，其中，不同的信号处理方式对应不同的解码器；a decoder determination submodule, configured to determine a target decoder corresponding to the first target audio data according to the signal processing method, wherein different signal processing methods correspond to different decoders;
置信度确定子模块，被配置为将该第一目标音频数据输入该目标解码器进行解码处理，并输出该第一目标音频数据的第一置信度。The confidence determination submodule is configured to input the first target audio data into the target decoder for decoding processing, and output a first confidence of the first target audio data.
可选地，该解码器确定子模块，被配置为：Optionally, the decoder determines the submodule, configured as follows:
通过预先设置的解码器关联关系，从多个解码器中确定该信号处理方式对应的解码器，该解码器关联关系包括不同的信号处理方式和该解码器的对应关系；Determine a decoder corresponding to the signal processing method from multiple decoders through a preset decoder association relationship, wherein the decoder association relationship includes a correspondence between different signal processing methods and the decoder;
将该信号处理方式对应的解码器作为该目标解码器。The decoder corresponding to the signal processing method is used as the target decoder.
可选地，该唤醒模块405包括：Optionally, the wake-up module 405 includes:
唤醒处理子模块，被配置为在每获取到一个目标解码器输出的该第一置信度的情况下，根据该目标解码器输出的第一目标置信度执行以下唤醒处理方式，直至唤醒该终端，或者，根据多个目标解码器输出的该第一置信度均执行了该唤醒处理方式；The wake-up processing submodule is configured to, when each time the first confidence output by a target decoder is obtained, perform the following wake-up processing method according to the first target confidence output by the target decoder until the terminal is woken up, or perform the wake-up processing method according to the first confidence output by multiple target decoders;
该唤醒处理方式包括：The wake-up processing method includes:
根据该第二置信度和该目标解码器输出的第一置信度确定是否唤醒该终端。Determine whether to wake up the terminal according to the second confidence and the first confidence output by the target decoder.
可选地，该唤醒模块包括：Optionally, the wake-up module includes:
目标置信度确定子模块，被配置为从该第二置信度中确定目标置信度，该目标置信度和该第一置信度是通过相同解码器解码得到的置信度；A target confidence determination submodule is configured to determine a target confidence from the second confidence, the target confidence and the first confidence being confidences obtained by decoding with the same decoder;
权重值确定子模块，被配置为根据该目标置信度和第三置信度，获取该第一置信度对应的权重值，该第三置信度包括该第二置信度中除该目标置信度之外的其他置信度；A weight value determination submodule is configured to obtain a weight value corresponding to the first confidence level according to the target confidence level and a third confidence level, wherein the third confidence level includes other confidence levels in the second confidence level except the target confidence level;
最终置信度获取子模块，被配置为根据该权重值和该第一置信度，获取最终置信度；A final confidence acquisition submodule is configured to acquire a final confidence according to the weight value and the first confidence;
唤醒子模块，被配置为根据该最终置信度确定是否唤醒该终端。The wake-up submodule is configured to determine whether to wake up the terminal according to the final confidence level.
可选地，该权重值确定子模块，还被配置为：Optionally, the weight value determination submodule is further configured as:
获取所述目标置信度和所述第三置信度之间的置信度差值；Obtaining a confidence difference between the target confidence and the third confidence;
根据所述置信度差值以及预设的对应关系，获取所述第一置信度对应的权重值。According to the confidence difference and a preset corresponding relationship, a weight value corresponding to the first confidence is obtained.
可选地，该唤醒子模块，还被配置为：Optionally, the wake-up submodule is further configured as:
在该最终置信度大于或等于预设置信度阈值的情况下，确定唤醒该终端。When the final confidence level is greater than or equal to a preset confidence threshold, it is determined to wake up the terminal.
可选地，该信号处理方式包括盲源分离或噪声抑制。Optionally, the signal processing method includes blind source separation or noise suppression.
通过上述装置，通过分别对目标时间段内采集的每路第一音频数据进行信号处理，得到多个第一目标音频数据，并分别获取该多个第一目标音频数据的第一置信度，之后，可以获取历史时间段内第二目标音频数据的第二置信度，并根据该第一置信度和该第二置信度确定是否唤醒该终端。也就是说，本公开可以根据目标时间段的第一置信度和历史时间段的第二置信度确定是否唤醒终端，这样，可以减少终端被误唤醒或漏唤醒的概率，从而可以提高语音识别系统的准确率。Through the above-mentioned device, by performing signal processing on each channel of first audio data collected in the target time period, multiple first target audio data are obtained, and the first confidence of the multiple first target audio data is obtained respectively. After that, the second confidence of the second target audio data in the historical time period can be obtained, and whether to wake up the terminal is determined according to the first confidence and the second confidence. In other words, the present disclosure can determine whether to wake up the terminal according to the first confidence of the target time period and the second confidence of the historical time period. In this way, the probability of the terminal being woken up by mistake or missed can be reduced, thereby improving the accuracy of the speech recognition system.
关于上述实施例中的装置，其中各个模块执行操作的具体方式已经在有关该方法的实施例中进行了详细描述，此处将不做详细阐述说明。Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.
本公开还提供一种计算机可读存储介质，其上存储有计算机程序指令，该程序指令被处理器执行时实现本公开提供的唤醒控制方法的步骤。The present disclosure also provides a computer-readable storage medium on which computer program instructions are stored. When the program instructions are executed by a processor, the steps of the wake-up control method provided by the present disclosure are implemented.
图5是根据一示例性实施例示出的一种终端500的框图。例如，终端500可以是移动电话，计算机，数字广播终端，消息收发设备，游戏控制台，平板设备，医疗设备，健身设备，个人数字助理等。Fig. 5 is a block diagram of a terminal 500 according to an exemplary embodiment. For example, the terminal 500 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.
参照图5，终端500可以包括以下一个或多个组件：处理组件502，存储器504，电力组件506，多媒体组件508，音频组件510，输入/输出(I/O)的接口512，传感器组件514，以及通信组件516。5 , terminal 500 may include one or more of the following components: a processing component 502 , a memory 504 , a power component 506 , a multimedia component 508 , an audio component 510 , an input/output (I/O) interface 512 , a sensor component 514 , and a communication component 516 .
处理组件502通常控制终端500的整体操作，诸如与显示，电话呼叫，数据通信，相机操作和记录操作相关联的操作。处理组件502可以包括一个或多个处理器520来执行指令，以完成上述的唤醒控制方法的全部或部分步骤。此外，处理组件502可以包括一个或多个模块，便于处理组件502和其他组件之间的交互。例如，处理组件502可以包括多媒体模块，以方便多媒体组件508和处理组件502之间的交互。The processing component 502 generally controls the overall operation of the terminal 500, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 502 may include one or more processors 520 to execute instructions to complete all or part of the steps of the above-mentioned wake-up control method. In addition, the processing component 502 may include one or more modules to facilitate the interaction between the processing component 502 and other components. For example, the processing component 502 may include a multimedia module to facilitate the interaction between the multimedia component 508 and the processing component 502.
存储器504被配置为存储各种类型的数据以支持在终端500的操作。这些数据的示例包括用于在终端500上操作的任何应用程序或方法的指令，联系人数据，电话簿数据，消息，图片，视频等。存储器504可以由任何类型的易失性或非易失性存储设备或者它们的组合实现，如静态随机存取存储器(SRAM)，电可擦除可编程只读存储器(EEPROM)，可擦除可编程只读存储器(EPROM)，可编程只读存储器(PROM)，只读存储器(ROM)，磁存储器，快闪存储器，磁盘或光盘。The memory 504 is configured to store various types of data to support operations at the terminal 500. Examples of such data include instructions for any application or method operating on the terminal 500, contact data, phone book data, messages, pictures, videos, etc. The memory 504 may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk, or an optical disk.
电力组件506为终端500的各种组件提供电力。电力组件506可以包括电源管理系统，一个或多个电源，及其他与为终端500生成、管理和分配电力相关联的组件。Power component 506 provides power to various components of terminal 500. Power component 506 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to terminal 500.
多媒体组件508包括在所述终端500和用户之间的提供一个输出接口的屏幕。在一些实施例中，屏幕可以包括液晶显示器(LCD)和触摸面板(TP)。如果屏幕包括触摸面板，屏幕可以被实现为触摸屏，以接收来自用户的输入信号。触摸面板包括一个或多个触摸传感器以感测触摸、滑动和触摸面板上的手势。所述触摸传感器可以不仅感测触摸或滑动动作的边界，而且还检测与所述触摸或滑动操作相关的持续时间和压力。在一些实施例中，多媒体组件508包括一个前置摄像头和/或后置摄像头。当终端500处于操作模式，如拍摄模式或视频模式时，前置摄像头和/或后置摄像头可以接收外部的多媒体数据。每个前置摄像头和后置摄像头可以是一个固定的光学透镜系统或具有焦距和光学变焦能力。The multimedia component 508 includes a screen that provides an output interface between the terminal 500 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 508 includes a front camera and/or a rear camera. When the terminal 500 is in an operating mode, such as a shooting mode or a video mode, the front camera and/or the rear camera may receive external multimedia data. Each front camera and the rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.
音频组件510被配置为输出和/或输入音频信号。例如，音频组件510包括一个麦克风(MIC)，当终端500处于操作模式，如呼叫模式、记录模式和语音识别模式时，麦克风被配置为接收外部音频信号。所接收的音频信号可以被进一步存储在存储器504或经由通信组件516发送。在一些实施例中，音频组件510还包括一个扬声器，用于输出音频信号。The audio component 510 is configured to output and/or input audio signals. For example, the audio component 510 includes a microphone (MIC), and when the terminal 500 is in an operation mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 504 or sent via the communication component 516. In some embodiments, the audio component 510 also includes a speaker for outputting audio signals.
I/O接口512为处理组件502和外围接口模块之间提供接口，上述外围接口模块可以是键盘，点击轮，按钮等。这些按钮可包括但不限于：主页按钮、音量按钮、启动按钮和锁定按钮。I/O interface 512 provides an interface between processing component 502 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.
传感器组件514包括一个或多个传感器，用于为终端500提供各个方面的状态评估。例如，传感器组件514可以检测到终端500的打开/关闭状态，组件的相对定位，例如所述组件为终端500的显示器和小键盘，传感器组件514还可以检测终端500或终端500一个组件的位置改变，用户与终端500接触的存在或不存在，终端500方位或加速/减速和终端500的温度变化。传感器组件514可以包括接近传感器，被配置用来在没有任何的物理接触时检测附近物体的存在。传感器组件514还可以包括光传感器，如CMOS或CCD图像传感器，用于在成像应用中使用。在一些实施例中，该传感器组件514还可以包括加速度传感器，陀螺仪传感器，磁传感器，压力传感器或温度传感器。The sensor assembly 514 includes one or more sensors for providing various aspects of status assessment for the terminal 500. For example, the sensor assembly 514 can detect the open/closed state of the terminal 500, the relative positioning of the components, such as the display and keypad of the terminal 500, and the sensor assembly 514 can also detect the position change of the terminal 500 or a component of the terminal 500, the presence or absence of contact between the user and the terminal 500, the orientation or acceleration/deceleration of the terminal 500, and the temperature change of the terminal 500. The sensor assembly 514 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 514 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 514 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
通信组件516被配置为便于终端500和其他设备之间有线或无线方式的通信。终端500可以接入基于通信标准的无线网络，如WiFi，2G或3G，或它们的组合。在一个示例性实施例中，通信组件516经由广播信道接收来自外部广播管理系统的广播信号或广播相关信息。在一个示例性实施例中，所述通信组件516还包括近场通信(NFC)模块，以促进短程通信。例如，在NFC模块可基于射频识别(RFID)技术，红外数据协会(IrDA)技术，超宽带(UWB)技术，蓝牙(BT)技术和其他技术来实现。The communication component 516 is configured to facilitate wired or wireless communication between the terminal 500 and other devices. The terminal 500 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 516 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 516 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.
在示例性实施例中，终端500可以被一个或多个应用专用集成电路(ASIC)、数字信号处理器(DSP)、数字信号处理设备(DSPD)、可编程逻辑器件(PLD)、现场可编程门阵列(FPGA)、控制器、微控制器、微处理器或其他电子元件实现，用于执行上述唤醒控制方法。In an exemplary embodiment, the terminal 500 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to perform the above-mentioned wake-up control method.
在示例性实施例中，还提供了一种包括指令的非临时性计算机可读存储介质，例如包括指令的存储器504，上述指令可由终端500的处理器520执行以完成上述唤醒控制方法。例如，所述非临时性计算机可读存储介质可以是ROM、随机存取存储器(RAM)、CD-ROM、磁带、软盘和光数据存储设备等。In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 504 including instructions, and the instructions can be executed by a processor 520 of a terminal 500 to complete the above-mentioned wake-up control method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.
在另一示例性实施例中，还提供一种计算机程序产品，该计算机程序产品包含能够由可编程的装置执行的计算机程序，该计算机程序具有当由该可编程的装置执行时用于执行上述的唤醒控制方法的代码部分。In another exemplary embodiment, a computer program product is further provided. The computer program product includes a computer program executable by a programmable device. The computer program has a code portion for executing the above-mentioned wake-up control method when executed by the programmable device.
本领域技术人员在考虑说明书及实践本公开后，将容易想到本公开的其它实施方案。本申请旨在涵盖本公开的任何变型、用途或者适应性变化，这些变型、用途或者适应性变化遵循本公开的一般性原理并包括本公开未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的，本公开的真正范围和精神由下面的权利要求指出。Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the present disclosure. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.
应当理解的是，本公开并不局限于上面已经描述并在附图中示出的精确结构，并且可以在不脱离其范围进行各种修改和改变。本公开的范围仅由所附的权利要求来限制。It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011303745.7ACN112509596B (en) | 2020-11-19 | 2020-11-19 | Wakeup control method, wakeup control device, storage medium and terminal |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011303745.7ACN112509596B (en) | 2020-11-19 | 2020-11-19 | Wakeup control method, wakeup control device, storage medium and terminal |
| Publication Number | Publication Date |
|---|---|
| CN112509596A CN112509596A (en) | 2021-03-16 |
| CN112509596Btrue CN112509596B (en) | 2024-07-09 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011303745.7AActiveCN112509596B (en) | 2020-11-19 | 2020-11-19 | Wakeup control method, wakeup control device, storage medium and terminal |
| Country | Link |
|---|---|
| CN (1) | CN112509596B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114229637B (en)* | 2021-12-03 | 2024-02-27 | 北京声智科技有限公司 | Elevator floor determination method, device, equipment and computer readable storage medium |
| CN115050013A (en)* | 2022-06-14 | 2022-09-13 | 南京人工智能高等研究院有限公司 | Behavior detection method and device, vehicle, storage medium and electronic equipment |
| CN115472161B (en)* | 2022-07-27 | 2025-05-02 | 北京声智科技有限公司 | Voice wake-up method, device, equipment and storage medium |
| CN115079587B (en)* | 2022-08-04 | 2025-07-08 | 四川长虹空调有限公司 | Intelligent household appliance control method, intelligent household appliance control device, computer equipment and storage medium |
| CN115687685B (en)* | 2022-11-15 | 2025-08-22 | 北京云迹科技股份有限公司 | Data processing method and device |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111508493A (en)* | 2020-04-20 | 2020-08-07 | Oppo广东移动通信有限公司 | Voice wake-up method, device, electronic device and storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI639153B (en)* | 2015-11-03 | 2018-10-21 | 絡達科技股份有限公司 | Electronic apparatus and voice trigger method therefor |
| CN106653010B (en)* | 2015-11-03 | 2020-07-24 | 络达科技股份有限公司 | Electronic device and method for waking up through voice recognition |
| CN105654949B (en)* | 2016-01-07 | 2019-05-07 | 北京云知声信息技术有限公司 | A kind of voice awakening method and device |
| TWI682385B (en)* | 2018-03-16 | 2020-01-11 | 緯創資通股份有限公司 | Speech service control apparatus and method thereof |
| CN110047485B (en)* | 2019-05-16 | 2021-09-28 | 北京地平线机器人技术研发有限公司 | Method and apparatus for recognizing wake-up word, medium, and device |
| CN110428810B (en)* | 2019-08-30 | 2020-10-30 | 北京声智科技有限公司 | Voice wake-up recognition method and device and electronic equipment |
| CN110838306B (en)* | 2019-11-12 | 2022-05-13 | 广州视源电子科技股份有限公司 | Voice signal detection method, computer storage medium and related equipment |
| CN111522592A (en)* | 2020-04-24 | 2020-08-11 | 腾讯科技（深圳）有限公司 | Intelligent terminal awakening method and device based on artificial intelligence |
| CN111696562B (en)* | 2020-04-29 | 2022-08-19 | 华为技术有限公司 | Voice wake-up method, device and storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111508493A (en)* | 2020-04-20 | 2020-08-07 | Oppo广东移动通信有限公司 | Voice wake-up method, device, electronic device and storage medium |
| Publication number | Publication date |
|---|---|
| CN112509596A (en) | 2021-03-16 |
| Publication | Publication Date | Title |
|---|---|---|
| CN112509596B (en) | Wakeup control method, wakeup control device, storage medium and terminal | |
| US10600415B2 (en) | Method, apparatus, device, and storage medium for voice interaction | |
| EP3933570A1 (en) | Method and apparatus for controlling a voice assistant, and computer-readable storage medium | |
| EP3136793A1 (en) | Method and apparatus for awakening electronic device | |
| EP3779968A1 (en) | Audio processing | |
| CN109087650B (en) | Voice wake-up method and device | |
| EP3133874A1 (en) | Method and apparatus for starting energy saving mode | |
| CN111696553B (en) | Voice processing method, device and readable medium | |
| KR20150069535A (en) | Method for performing voice control operation on terminal, and apparatus | |
| EP3024211A1 (en) | Method and device for announcing voice call | |
| CN111063354A (en) | Man-machine interaction method and device | |
| CN109599104A (en) | Multi-beam choosing method and device | |
| CN111968680B (en) | Voice processing method, device and storage medium | |
| CN109788367A (en) | A kind of information cuing method, device, electronic equipment and storage medium | |
| CN109243447A (en) | Voice sends triggering method and device | |
| US20170034347A1 (en) | Method and device for state notification and computer-readable storage medium | |
| CN112489653B (en) | Speech recognition method, device and storage medium | |
| CN112489650B (en) | Wakeup control method, wakeup control device, storage medium and terminal | |
| US11561278B2 (en) | Method and device for processing information based on radar waves, terminal, and storage medium | |
| CN109922203A (en) | Terminal puts out screen method and apparatus | |
| CN109119075A (en) | Speech recognition scene awakening method and device | |
| CN112866480B (en) | Information processing method, device, electronic device and storage medium | |
| US11862158B2 (en) | Method and apparatus for controlling device, and readable storage medium | |
| CN112863511B (en) | Signal processing method, device and storage medium | |
| CN107979695B (en) | Network message receiving method and device, and storage medium |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |