技术领域technical field
本公开的实施例涉及计算机技术领域，具体涉及用于输出信息的方法和装置。Embodiments of the present disclosure relate to the field of computer technology, and in particular, to a method and apparatus for outputting information.
背景技术Background technique
在文章自动生成领域，多媒体转写自动生成的文章还比较少，大多是根据结构化的文本数据来生成文章，这使得数据来源单一，生成的文章不够丰富、广泛；而人工编辑的多媒体文章又非常的耗时和繁琐，造成不必要的人力、财力的开销。常规的方法主要是人工编辑，通过人工将相关音频转化为文本，然后根据音频主题等在网络上查找相关图片，最后人工的将转化后的文本和图片渲染。In the field of automatic article generation, there are still relatively few articles automatically generated by multimedia transcription, and most of them are generated based on structured text data, which makes the data source single, and the generated articles are not rich and extensive. It is very time-consuming and cumbersome, resulting in unnecessary human and financial expenses. The conventional method is mainly manual editing, converting the relevant audio into text manually, then searching for relevant pictures on the Internet according to the audio theme, etc., and finally manually rendering the converted text and pictures.
基于人工的方法主要问题在于：(1)对于音频的转换：采用人工的方式费时费力，准确率也不一定高；(2)配图的选择：根据主题选择相关的图片，人工搜索的方式将耗费大量人力；(3)文章的组织渲染，将相关文本和图片组织最终生成一篇阅读性强的文章。The main problems of the manual-based method are: (1) For audio conversion: the manual method is time-consuming and labor-intensive, and the accuracy rate is not necessarily high; (2) The selection of pictures: select relevant pictures according to the theme, and the manual search method will It consumes a lot of manpower; (3) the organization and rendering of the article, organize the relevant text and pictures to finally generate a highly readable article.
发明内容SUMMARY OF THE INVENTION
本公开的实施例提出了用于输出信息的方法和装置。Embodiments of the present disclosure propose methods and apparatuses for outputting information.
第一方面，本公开的实施例提供了一种用于输出信息的方法，包括：获取待转换的音频信息；将音频信息转换成文本信息；将文本信息进行切词，得到词序列；对于词序列中的词，通过预先训练的词连接概率模型得到的词连接概率表查询该词和与该词的下一个词之间连接概率和该词与各类标点的连接概率，以及基于查询到的连接概率确定该词的连接目标；将词序列中各词与相应的连接目标连接起来生成带标点的文章进行输出。In a first aspect, an embodiment of the present disclosure provides a method for outputting information, including: acquiring audio information to be converted; converting the audio information into text information; segmenting the text information to obtain a word sequence; For a word in the sequence, query the connection probability between the word and the next word of the word and the connection probability between the word and various punctuation points through the word connection probability table obtained by the pre-trained word connection probability model, and based on the query. The connection probability determines the connection target of the word; each word in the word sequence is connected with the corresponding connection target to generate a punctuated article for output.
在一些实施例中，词连接概率表是通过以下步骤得到的：获取训练样本集合，训练样本包括含有标点的句子；将训练样本集合中的训练样本的句子作为LSTM模型的输入，训练得到词连接概率模型；根据词连接概率模型训练的中间过程中得到每个词与词之间的概率和每个词与各标点之间的概率生成词连接概率表。In some embodiments, the word connection probability table is obtained through the following steps: obtaining a training sample set, the training samples include sentences containing punctuation; using the sentences of the training samples in the training sample set as the input of the LSTM model, and training to obtain word connections Probability model; according to the intermediate process of word connection probability model training, the probability between each word and the probability between each word and each punctuation point is obtained to generate a word connection probability table.
在一些实施例中，获取训练样本集合，包括：获取样本文章，将样本文章按一个大句的粒度进行切分得到样本句集合，其中，大句是指以句号、问号或感叹号结尾的句子；对于样本句集合中的样本句，将该句进行切词后生成词向量作为训练样本。In some embodiments, acquiring a training sample set includes: acquiring a sample article, and dividing the sample article according to the granularity of a large sentence to obtain a sample sentence set, wherein a large sentence refers to a sentence ending with a period, a question mark or an exclamation mark; For the sample sentence in the sample sentence set, the word vector is generated after the sentence is segmented as a training sample.
在一些实施例中，该方法还包括：将文章分成至少一个段落。In some embodiments, the method further includes dividing the article into at least one paragraph.
在一些实施例中，该方法还包括：确定文章的主题和实体；获取与文章的主题和实体匹配的图像；根据图像和文章生成图文信息。In some embodiments, the method further includes: determining the subject and entity of the article; acquiring an image matching the subject and entity of the article; generating graphic and textual information according to the image and the article.
在一些实施例中，该方法还包括：将图文信息进行排版优化。In some embodiments, the method further includes: performing layout optimization on the graphic and text information.
第二方面，本公开的实施例提供了一种用于输出信息的装置，包括：获取单元，被配置成获取待转换的音频信息；转换单元，被配置成将音频信息转换成文本信息；切词单元，被配置成将文本信息进行切词，得到词序列；判断单元，被配置成对于词序列中的词，通过预先训练的词连接概率模型得到的词连接概率表查询该词和与该词的下一个词之间连接概率和该词与各类标点的连接概率，以及基于查询到的连接概率确定该词的连接目标；连接单元，被配置成将词序列中各词与相应的连接目标连接起来生成带标点的文章进行输出。In a second aspect, an embodiment of the present disclosure provides an apparatus for outputting information, comprising: an acquisition unit configured to acquire audio information to be converted; a conversion unit configured to convert the audio information into text information; The word unit is configured to segment the text information to obtain a word sequence; the judgment unit is configured to query the word and the word connection probability table obtained by the pre-trained word connection probability model for the words in the word sequence. The connection probability between the next word of the word and the connection probability between the word and various punctuation marks, and the connection target of the word is determined based on the queried connection probability; the connection unit is configured to connect each word in the word sequence with the corresponding connection The targets are concatenated to generate punctuated articles for output.
在一些实施例中，该装置还包括训练单元，被配置成：获取训练样本集合，训练样本包括含有标点的句子；将训练样本集合中的训练样本的句子作为LSTM模型的输入，训练得到词连接概率模型；根据词连接概率模型训练的中间过程中得到每个词与词之间的概率和每个词与各标点之间的概率生成词连接概率表。In some embodiments, the apparatus further includes a training unit configured to: obtain a training sample set, the training samples include sentences containing punctuation; use the sentences of the training samples in the training sample set as the input of the LSTM model, and train to obtain word connections Probability model; according to the intermediate process of word connection probability model training, the probability between each word and the probability between each word and each punctuation point is obtained to generate a word connection probability table.
在一些实施例中，训练单元进一步被配置成：获取样本文章，将样本文章按一个大句的粒度进行切分得到样本句集合，其中，大句是指以句号、问号或感叹号结尾的句子；对于样本句集合中的样本句，将该句进行切词后生成词向量作为训练样本。In some embodiments, the training unit is further configured to: acquire a sample article, and segment the sample article according to the granularity of a large sentence to obtain a sample sentence set, wherein a large sentence refers to a sentence ending with a period, a question mark or an exclamation mark; For the sample sentence in the sample sentence set, the word vector is generated after the sentence is segmented as a training sample.
在一些实施例中，该装置还包括分段单元，被配置成：将文章分成至少一个段落。In some embodiments, the apparatus further includes a segmentation unit configured to: divide the article into at least one paragraph.
在一些实施例中，该装置还包括配图单元，被配置成：确定文章的主题和实体；获取与文章的主题和实体匹配的图像；根据图像和文章生成图文信息。In some embodiments, the apparatus further includes an image matching unit configured to: determine the subject and entity of the article; acquire an image matching the subject and entity of the article; and generate graphic and textual information according to the image and the article.
在一些实施例中，该装置还包括排版单元，被配置成：将图文信息进行排版优化。In some embodiments, the apparatus further includes a typesetting unit configured to: optimize the typesetting of the graphic and text information.
第三方面，本公开的实施例提供了一种电子设备，包括：一个或多个处理器；存储装置，其上存储有一个或多个程序，当一个或多个程序被一个或多个处理器执行，使得一个或多个处理器实现如第一方面中任一的方法。In a third aspect, embodiments of the present disclosure provide an electronic device, including: one or more processors; and a storage device on which one or more programs are stored, when the one or more programs are processed by one or more The processor executes such that the one or more processors implement a method as in any one of the first aspects.
第四方面，本公开的实施例提供了一种计算机可读介质，其上存储有计算机程序，其中，程序被处理器执行时实现如第一方面中任一的方法。In a fourth aspect, embodiments of the present disclosure provide a computer-readable medium on which a computer program is stored, wherein the program implements the method according to any one of the first aspects when executed by a processor.
本公开的实施例提供的用于输出信息的方法和装置，可以根据音频解析出的文本内容进行句子链接并分段，然后根据本文主题内容进行配图，最后对文本、图片进行排版优化生成文章。相比于传统文章生成系统，该系统的数据更丰富、多样，来源也更广泛。相比传统的小编手写文章，有更高的时效性和覆盖度，同时也节省了人力成本和时间成本。The method and device for outputting information provided by the embodiments of the present disclosure can link and segment sentences according to the text content parsed from the audio, then arrange pictures according to the subject content of the article, and finally optimize the typesetting of the text and pictures to generate an article . Compared with traditional article generation systems, the data of this system is richer, more diverse, and comes from a wider range of sources. Compared with traditional handwritten articles, it has higher timeliness and coverage, and also saves labor costs and time costs.
附图说明Description of drawings
通过阅读参照以下附图所作的对非限制性实施例所作的详细描述，本公开的其它特征、目的和优点将会变得更明显：Other features, objects and advantages of the present disclosure will become more apparent upon reading the detailed description of non-limiting embodiments taken with reference to the following drawings:
图1是本公开的一个实施例可以应用于其中的示例性系统架构图；FIG. 1 is an exemplary system architecture diagram to which an embodiment of the present disclosure may be applied;
图2是根据本公开的用于输出信息的方法的一个实施例的流程图；2 is a flowchart of one embodiment of a method for outputting information according to the present disclosure;
图3是根据本公开的用于输出信息的方法的一个应用场景的示意图；3 is a schematic diagram of an application scenario of the method for outputting information according to the present disclosure;
图4是根据本公开的用于输出信息的方法的又一个实施例的流程图；4 is a flowchart of yet another embodiment of a method for outputting information according to the present disclosure;
图5a、5b是根据本公开的用于输出信息的方法的LSTM模型的网络结构示意图。5a and 5b are schematic diagrams of the network structure of the LSTM model of the method for outputting information according to the present disclosure.
图6是根据本公开的用于输出信息的装置的一个实施例的结构示意图；6 is a schematic structural diagram of an embodiment of an apparatus for outputting information according to the present disclosure;
图7是适于用来实现本公开的实施例的电子设备的计算机系统的结构示意图。7 is a schematic structural diagram of a computer system suitable for implementing an electronic device of an embodiment of the present disclosure.
具体实施方式Detailed ways
下面结合附图和实施例对本公开作进一步的详细说明。可以理解的是，此处所描述的具体实施例仅仅用于解释相关发明，而非对该发明的限定。另外还需要说明的是，为了便于描述，附图中仅示出了与有关发明相关的部分。The present disclosure will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the related invention, but not to limit the invention. In addition, it should be noted that, for the convenience of description, only the parts related to the related invention are shown in the drawings.
需要说明的是，在不冲突的情况下，本公开中的实施例及实施例中的特征可以相互组合。下面将参考附图并结合实施例来详细说明本公开。It should be noted that the embodiments of the present disclosure and the features of the embodiments may be combined with each other under the condition of no conflict. The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.
图1示出了可以应用本公开的用于输出信息的方法或用于输出信息的装置的实施例的示例性系统架构100。FIG. 1 illustrates an exemplary system architecture 100 to which an embodiment of a method for outputting information or an apparatus for outputting information of the present disclosure may be applied.
如图1所示，系统架构100可以包括终端设备101、102、103，网络104和服务器105。网络104用以在终端设备101、102、103和服务器105之间提供通信链路的介质。网络104可以包括各种连接类型，例如有线、无线通信链路或者光纤电缆等等。As shown in FIG. 1 , the system architecture 100 may include terminal devices 101 , 102 , and 103 , a network 104 and a server 105 . The network 104 is a medium used to provide a communication link between the terminal devices 101 , 102 , 103 and the server 105 . The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, among others.
用户可以使用终端设备101、102、103通过网络104与服务器105交互，以接收或发送消息等。终端设备101、102、103上可以安装有各种通讯客户端应用，例如音频转文字应用、网页浏览器应用、购物类应用、搜索类应用、即时通信工具、邮箱客户端、社交平台软件等。The user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages and the like. Various communication client applications can be installed on the terminal devices 101 , 102 and 103 , such as audio-to-text applications, web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, and the like.
终端设备101、102、103可以是硬件，也可以是软件。当终端设备101、102、103为硬件时，可以是具有麦克风、显示屏并且支持音频转文字的各种电子设备，包括但不限于智能手机、平板电脑、电子书阅读器、MP3播放器(Moving Picture Experts Group Audio LayerIII，动态影像专家压缩标准音频层面3)、MP4(Moving Picture Experts Group AudioLayer IV，动态影像专家压缩标准音频层面4)播放器、膝上型便携计算机和台式计算机等等。当终端设备101、102、103为软件时，可以安装在上述所列举的电子设备中。其可以实现成多个软件或软件模块(例如用来提供分布式服务)，也可以实现成单个软件或软件模块。在此不做具体限定。The terminal devices 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, and 103 are hardware, they can be various electronic devices with microphones, display screens, and support for audio-to-text conversion, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group AudioLayer IV, Moving Picture Experts Group Audio Layer 4) Players, Laptops and Desktops, etc. When the terminal devices 101, 102, and 103 are software, they can be installed in the electronic devices listed above. It can be implemented as multiple software or software modules (eg, to provide distributed services), or as a single software or software module. There is no specific limitation here.
服务器105可以是提供各种服务的服务器，例如对终端设备101、102、103上显示的文字提供支持的后台编辑服务器。后台编辑服务器可以对接收到的音频等数据进行分析等处理，并将处理结果(例如根据音频生成的文章)反馈给终端设备。The server 105 may be a server that provides various services, such as a background editing server that supports the text displayed on the terminal devices 101 , 102 and 103 . The background editing server can analyze and process the received audio and other data, and feed back the processing result (for example, an article generated according to the audio) to the terminal device.
需要说明的是，服务器可以是硬件，也可以是软件。当服务器为硬件时，可以实现成多个服务器组成的分布式服务器集群，也可以实现成单个服务器。当服务器为软件时，可以实现成多个软件或软件模块(例如用来提供分布式服务的多个软件或软件模块)，也可以实现成单个软件或软件模块。在此不做具体限定。It should be noted that the server may be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster composed of multiple servers, or can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, multiple software or software modules for providing distributed services), or can be implemented as a single software or software module. There is no specific limitation here.
需要说明的是，本公开的实施例所提供的用于输出信息的方法可以由终端设备101、102、103执行，也可以由服务器105执行。相应地，用于输出信息的装置可以设置于终端设备101、102、103中，也可以设置于服务器105中。在此不做具体限定。It should be noted that the methods for outputting information provided by the embodiments of the present disclosure may be executed by the terminal devices 101 , 102 , and 103 , or may be executed by the server 105 . Correspondingly, the means for outputting information may be provided in the terminal devices 101 , 102 , and 103 , or may be provided in the server 105 . There is no specific limitation here.
应该理解，图1中的终端设备、网络和服务器的数目仅仅是示意性的。根据实现需要，可以具有任意数目的终端设备、网络和服务器。It should be understood that the numbers of terminal devices, networks and servers in FIG. 1 are merely illustrative. There can be any number of terminal devices, networks and servers according to implementation needs.
继续参考图2，示出了根据本公开的用于输出信息的方法的一个实施例的流程200。该用于输出信息的方法，包括以下步骤：With continued reference to FIG. 2, a flow 200 of one embodiment of a method for outputting information according to the present disclosure is shown. The method for outputting information includes the following steps:
步骤201，获取待转换的音频信息。Step 201: Acquire audio information to be converted.
在本实施例中，用于输出信息的方法的执行主体(例如图1所示的服务器)可以通过有线连接方式或者无线连接方式从用户利用其进行语音写稿的终端接收音频信息。音频信息可以是各种格式的音频文件。其包含大量的语句。音频文件的名称中可包含这段音频的标题。In this embodiment, the execution body of the method for outputting information (for example, the server shown in FIG. 1 ) may receive audio information from a terminal through which the user performs voice writing through a wired connection or a wireless connection. The audio information can be audio files in various formats. It contains a large number of statements. The title of the audio file can be included in the name of the audio file.
步骤202，将音频信息转换成文本信息。Step 202, converting audio information into text information.
在本实施例中，可通过现有的自动语音识别(ASR，Automatic SpeechRecognition)技术将音频信息转换成整段文本。基于ASR解析得到的文本是一段没有断句的文本，所以还需要根据语意对其进行切割并链接，打上标点。In this embodiment, the audio information can be converted into a whole text by using the existing automatic speech recognition (ASR, Automatic Speech Recognition) technology. The text parsed based on ASR is a piece of text without sentence segmentation, so it needs to be cut and linked according to its semantics, and punctuated.
步骤203，将文本信息进行切词，得到词序列。In step 203, the text information is segmented into words to obtain a word sequence.
在本实施例中，将整段文本基于汉语或者英语的词法结构进行切词操作，得到整段音频的词序列。切词方法可包括最大逆向匹配法等常见切词方式。可先识别出音频的语种，例如，中文、英文或其它语种。然后根据该语种的词法结构进行切词操作。In this embodiment, a word segmentation operation is performed on the entire text based on the lexical structure of Chinese or English to obtain a word sequence of the entire audio. The word segmentation method can include common word segmentation methods such as maximum reverse matching method. The language of the audio can be identified first, for example, Chinese, English or other languages. Then the word segmentation operation is performed according to the lexical structure of the language.
步骤204，对于词序列中的词，通过预先训练的词连接概率模型得到的词连接概率表查询该词和与该词的下一个词之间连接概率和该词与各类标点的连接概率，以及基于查询到的连接概率确定该词的连接目标。Step 204, for the word in the word sequence, query the connection probability between the word and the next word of the word and the connection probability between the word and various punctuation points through the word connection probability table obtained by the pre-trained word connection probability model, and determining the connection target of the word based on the queried connection probability.
在本实施例中，根据词连接概率模型生成的词连接概率表，针对每个词计算其和下一个词及各类标点的概率，取概率值最大的词或者标点进行链接。词连接概率表用于表征词与词或各类标点的概率。我们将每个词都看作独立的，也就是说，每个词后边都有可能加上标点(，。？！；等)。针对当前词，分别计算该词与各种标点以及下一个词的概率，最后取概率最高的词进行链接。如果概率最高是下一个词，说明这里还不需要标点连接，直接进行词追加就好。如果概率最高的是标点，则在该词后边追加标点符号。对所有词进行如上步骤，最后得到用标点连接好的句子。例如，词序列“我”“爱”“中国”“因为”，依次查询“我”与“爱”之间的连接概率，以及“我”与句号、逗号等标点的连接概率。可得到“我”与“爱”之间的连接概率最大，因此“我”“爱”之间不使用标点。而“中国”与句号的连接概率远大于“中国”与“因为”，也大于“中国”与其它标点的连接概率，因此在“中国”后面加上句号。词连接概率模型是本子系统的一个重点，需要训练相关模型，得到词与词之间出现的概率从而生成词连接概率表，然后取概率最高的词作为该词的连接目标。词连接概率表的生成过程将在步骤401-403中介绍。In this embodiment, according to the word connection probability table generated by the word connection probability model, the probability of each word and the next word and various punctuations is calculated, and the word or punctuation with the highest probability value is selected for linking. The word connection probability table is used to represent the probability of words and words or various types of punctuation. We treat each word as independent, that is, each word may be followed by punctuation (, .?!; etc.). For the current word, calculate the probability of the word, various punctuation marks and the next word, and finally take the word with the highest probability to link. If the highest probability is the next word, it means that there is no need for punctuation connection here, just add the word directly. If the highest probability is punctuation, append punctuation to the word. Perform the above steps for all words, and finally get a sentence connected with punctuation. For example, the word sequence "I", "love", "China", "because", query the connection probability between "I" and "love" in turn, and the connection probability between "I" and punctuation such as periods and commas. It can be obtained that the connection probability between "me" and "love" is the highest, so no punctuation is used between "me" and "love". The connection probability between "China" and a period is much greater than that between "China" and "because", and also greater than the connection probability between "China" and other punctuation marks, so a period is added after "China". The word connection probability model is a key point of this subsystem. It is necessary to train related models to obtain the probability of occurrence between words to generate a word connection probability table, and then take the word with the highest probability as the connection target of the word. The generation process of the word connection probability table will be introduced in steps 401-403.
步骤205，将词序列中各词与相应的连接目标连接起来生成带标点的文章进行输出。Step 205: Connect each word in the word sequence with the corresponding connection target to generate an article with punctuation for output.
在本实施例中，根据步骤204的结果，将词序列中各词与相应的连接目标连接起来生成带标点的文章进行输出。In this embodiment, according to the result of step 204, each word in the word sequence is connected with the corresponding connection target to generate an article with punctuation for output.
在本实施例的一些可选的实现方式中，该方法还包括：将文章分成至少一个段落。可对文章进行语义分析，然后根据语义将文章分段。相同语义的文字内容归为一段。In some optional implementations of this embodiment, the method further includes: dividing the article into at least one paragraph. Articles can be semantically analyzed and then segmented according to the semantics. Text content with the same semantics is grouped into a paragraph.
在本实施例的一些可选的实现方式中，该方法还包括：确定文章的主题和实体；获取与文章的主题和实体匹配的图像；根据图像和文章生成图文信息。根据音频转换模块得到的文本数据，挖掘出文本中的实体(这里实体是比较细粒度的，包括人物比如明星、事物比如银行)以及主题(财经、娱乐、体育等类别)，然后根据实体去实体图库检索相关实体图、根据主题去主题图库检索相关主题图。这些图片即是文本相关图片，可以直接用做文章的配图。In some optional implementations of this embodiment, the method further includes: determining the subject and entity of the article; acquiring an image matching the subject and entity of the article; generating graphic and textual information according to the image and the article. According to the text data obtained by the audio conversion module, the entities in the text (here entities are relatively fine-grained, including characters such as stars, things such as banks) and themes (financial, entertainment, sports and other categories) are mined in the text, and then the entities are removed according to the entities. Retrieve related entity graphs from the gallery, and go to the subject gallery to retrieve related subject graphs according to the subject. These pictures are text-related pictures, which can be directly used as pictures for articles.
在本实施例的一些可选的实现方式中，该方法还包括：将图文信息进行排版优化。自动将图片插入到文章比较合理的位置，并调整图片尺寸，使得文字内容和图片的面积比例达到预定值。In some optional implementations of this embodiment, the method further includes: performing typesetting optimization on the graphic and text information. Automatically insert the picture into the reasonable position of the article, and adjust the size of the picture, so that the area ratio of the text content and the picture reaches the predetermined value.
继续参见图3，图3是根据本实施例的用于输出信息的方法的应用场景的一个示意图。在图3的应用场景中，服务器接收到终端发送的音频文件“大理风光”。用户在该音频文件中用语音描述了云南大理的风土人情。通过ASR技术，将音频文件解析成整段的文本。然后将整段的文本切词后，查询词与词之间、词与各标点之间的连接概率，将概率最大的词或标点作为连接目标。每个词进行连接后生成带标点的文章。还可根据文本内容进行图片检索，找到合适的配图。然后根据文本语义将文章分段。最后将搜索到的图片插入到文章中再进行润色处理。Continue to refer to FIG. 3 , which is a schematic diagram of an application scenario of the method for outputting information according to this embodiment. In the application scenario of FIG. 3 , the server receives the audio file "Dali Scenery" sent by the terminal. In this audio file, the user described the local customs and conditions of Dali, Yunnan by voice. Through ASR technology, the audio file is parsed into an entire text. Then, after segmenting the entire text into words, query the connection probability between words and words and between words and punctuation, and take the word or punctuation with the highest probability as the connection target. After each word is concatenated, a punctuated article is generated. You can also search for pictures according to the text content to find suitable pictures. Articles are then segmented according to textual semantics. Finally, insert the searched pictures into the article and then polish them.
本公开的上述实施例提供的方法，可以根据音频解析出的文本内容进行句子链接&分段，然后根据本文主题内容进行配图，最后对文本、图片进行排版优化生成文章。相比于传统文章生成系统，该系统的数据更丰富、多样，来源也更广泛；相比传统的小编手写文章，有更高的时效性和覆盖度，同时也节省了人力成本和时间成本。The method provided by the above embodiments of the present disclosure can perform sentence linking & segmentation according to the text content parsed from the audio, then map according to the subject content of the text, and finally optimize the typesetting of the text and pictures to generate an article. Compared with the traditional article generation system, the data of this system is richer and more diverse, and the sources are also wider; compared with the traditional handwritten articles, it has higher timeliness and coverage, and also saves labor costs and time costs. .
进一步参考图4，其示出了用于输出信息的方法的又一个实施例的流程400。该用于输出信息的方法的流程400，包括以下步骤：With further reference to Figure 4, a flow 400 of yet another embodiment of a method for outputting information is shown. The process 400 of the method for outputting information includes the following steps:
步骤401，获取训练样本集合。Step 401, acquiring a training sample set.
在本实施例中，用于输出信息的方法的执行主体(例如图1所示的服务器)可以通过有线连接方式或者无线连接方式获取训练样本集合，其中，训练样本包括含有标点的句子。流程400的执行主体可与流程200的执行主体相同，也可是不同执行主体。可由第三方服务器执行流程400后生成词连接概率表，再发给流程200的执行主体使用。In this embodiment, the execution body of the method for outputting information (for example, the server shown in FIG. 1 ) may acquire a training sample set through a wired connection or a wireless connection, where the training samples include sentences containing punctuation. The execution body of the process 400 may be the same as the execution body of the process 200, or may be different execution bodies. The word connection probability table can be generated by the third-party server after executing the process 400, and then sent to the execution subject of the process 200 for use.
使用正常的新闻文本或者文章作为训练数据。Use normal news text or articles as training data.
首先，进行文章切句，将文章按照一个大句的粒度进行切分，大句是指以句号、问号、感叹号结尾的句子。每一个大句作为一条数据；First, the article is divided into sentences, and the article is divided according to the granularity of a large sentence. A large sentence refers to a sentence ending with a period, a question mark, and an exclamation mark. Each big sentence is used as a piece of data;
然后，进行句子切词，根据英文或中文的词法结构对句子进行切词；Then, perform sentence segmentation, and segment the sentence according to the lexical structure of English or Chinese;
最后，进行词encode(编码)，讲每一个单词做embedding(嵌入)得到每个句子embedding表示，即得到了训练样本。这里的词包括标点符号。Finally, perform word encode (encoding), and make embedding (embedding) for each word to obtain the embedding representation of each sentence, that is, the training sample is obtained. Words here include punctuation.
步骤402，将训练样本集合中的训练样本的句子作为LSTM模型的输入，训练得到词连接概率模型。In step 402, the sentences of the training samples in the training sample set are used as the input of the LSTM model, and the word connection probability model is obtained by training.
在本实施例中，LSTM(Long Short-Term Memory)是长短期记忆网络，是一种时间循环神经网络，适合于处理和预测时间序列中间隔和延迟相对较长的重要事件。原始RNN的隐藏层只有一个状态(图5a)，它对于短期的输入非常敏感。那么，假如我们再增加一个状态(图5b)，让它来保存长期的状态。In this embodiment, LSTM (Long Short-Term Memory) is a long short-term memory network, which is a time-recurrent neural network, suitable for processing and predicting important events with relatively long intervals and delays in a time series. The hidden layer of the original RNN has only one state (Fig. 5a), which is very sensitive to short-term inputs. Then, suppose we add one more state (Figure 5b) and let it hold the long-term state.
LSTM同样是这样的结构，但是重复的模块拥有一个不同的结构。不同于单一神经网络层，这里是有四个，以一种非常特殊的方式进行交互。LSTMs have the same structure, but repeated modules have a different structure. Instead of a single neural network layer, here are four, interacting in a very specific way.
·在t时刻，LSTM的输入有三个：当前时刻网络的输入值、上一时刻LSTM的输出值、以及上一时刻的单元状态；LSTM的输出有两个：当前时刻LSTM输出值、和当前时刻的单元状态。At time t, there are three inputs to LSTM: the input value of the network at the current time, the output value of the LSTM at the previous time, and the unit state at the previous time; there are two outputs of the LSTM: the output value of the LSTM at the current time, and the current time. unit status.
·LSTM的关键，就是怎样控制长期状态。在这里，LSTM的思路是使用三个控制开关。第一个开关，负责控制继续保存长期状态；第二个开关，负责控制把即时状态输入到长期状态；第三个开关，负责控制是否把长期状态作为当前的LSTM的输出。The key to LSTM is how to control the long-term state. Here, the idea of LSTM is to use three control switches. The first switch is responsible for controlling the continued preservation of the long-term state; the second switch is responsible for controlling the input of the immediate state to the long-term state; the third switch is responsible for controlling whether to use the long-term state as the output of the current LSTM.
步骤403，根据词连接概率模型训练的中间过程中得到每个词与词之间的概率和每个词与各标点之间的概率生成词连接概率表。Step 403 , generating a word connection probability table according to the probability between each word and the probability between each word and each punctuation point in the intermediate process of the word connection probability model training.
在本实施例中，将embedding后的句子作为LSTM模型的输入，训练模型。拉取模型的中间过程，得到每个词与词之间的连接概率。将各词之间的连接概率进行统计分析得到词连接概率表。通过查词连接概率表，可得到词与词之前的连接概率。In this embodiment, the embedded sentence is used as the input of the LSTM model to train the model. Pull the intermediate process of the model to get the connection probability between each word and the word. Statistical analysis is performed on the connection probability between each word to obtain a word connection probability table. By looking up the word connection probability table, the connection probability between the word and the word can be obtained.
进一步参考图6，作为对上述各图所示方法的实现，本公开提供了一种用于输出信息的装置的一个实施例，该装置实施例与图2所示的方法实施例相对应，该装置具体可以应用于各种电子设备中。Referring further to FIG. 6 , as an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an apparatus for outputting information. The apparatus embodiment corresponds to the method embodiment shown in FIG. 2 . The device can be specifically applied to various electronic devices.
如图6所示，本实施例的用于输出信息的装置600包括：获取单元601、转换单元602、切词单元603、判断单元604和连接单元605。其中，获取单元601，被配置成获取待转换的音频信息；转换单元602，被配置成将音频信息转换成文本信息；切词单元603，被配置成将文本信息进行切词，得到词序列；判断单元604，被配置成对于词序列中的词，通过预先训练的词连接概率模型得到的词连接概率表查询该词和与该词的下一个词之间连接概率和该词与各类标点的连接概率，以及基于查询到的连接概率确定该词的连接目标；连接单元605，被配置成将词序列中各词与相应的连接目标连接起来生成带标点的文章进行输出。As shown in FIG. 6 , the apparatus 600 for outputting information in this embodiment includes: an acquisition unit 601 , a conversion unit 602 , a word segmentation unit 603 , a judgment unit 604 and a connection unit 605 . Wherein, the acquisition unit 601 is configured to acquire the audio information to be converted; the conversion unit 602 is configured to convert the audio information into text information; the word segmentation unit 603 is configured to segment the text information to obtain a word sequence; The judgment unit 604 is configured to query the word connection probability table obtained by the pre-trained word connection probability model for the word in the word sequence and the connection probability between the word and the next word of the word and the word and various punctuation points. and determine the connection target of the word based on the queried connection probability; the connection unit 605 is configured to connect each word in the word sequence with the corresponding connection target to generate an article with punctuation for output.
在本实施例中，用于输出信息的装置600的获取单元601、转换单元602、切词单元603、判断单元604和连接单元605的具体处理可以参考图2对应实施例中的步骤201、步骤202、步骤203、步骤204和步骤205。In this embodiment, for the specific processing of the acquiring unit 601 , the converting unit 602 , the word segmentation unit 603 , the determining unit 604 and the connecting unit 605 of the apparatus 600 for outputting information, reference may be made to steps 201 and 201 in the corresponding embodiment of FIG. 2 . 202 , step 203 , step 204 and step 205 .
在本实施例的一些可选的实现方式中，装置600还包括训练单元(附图中未示出)，被配置成：获取训练样本集合，其中，训练样本包括含有标点的句子；将训练样本集合中的训练样本的句子作为LSTM模型的输入，训练得到词连接概率模型；根据词连接概率模型训练的中间过程中得到每个词与词之间的概率和每个词与各标点之间的概率生成词连接概率表。In some optional implementations of this embodiment, the apparatus 600 further includes a training unit (not shown in the drawings), configured to: acquire a training sample set, wherein the training samples include sentences containing punctuation; The sentences of the training samples in the set are used as the input of the LSTM model, and the word connection probability model is obtained by training; according to the intermediate process of the word connection probability model training, the probability between each word and the word and the probability between each word and each punctuation are obtained. Probabilistically generated word join probability table.
在本实施例的一些可选的实现方式中，训练单元进一步被配置成：获取样本文章，将样本文章按一个大句的粒度进行切分得到样本句集合，其中，大句是指以句号、问号或感叹号结尾的句子；对于样本句集合中的样本句，将该句进行切词后生成词向量作为训练样本。In some optional implementation manners of this embodiment, the training unit is further configured to: acquire a sample article, and segment the sample article according to the granularity of a large sentence to obtain a sample sentence set, where a large sentence refers to a sentence with a period, A sentence ending with a question mark or an exclamation mark; for a sample sentence in the sample sentence set, segment the sentence to generate a word vector as a training sample.
在本实施例的一些可选的实现方式中，装置600还包括分段单元(附图中未示出)，被配置成：将文章分成至少一个段落。In some optional implementations of this embodiment, the apparatus 600 further includes a segmentation unit (not shown in the drawings) configured to: divide the article into at least one paragraph.
在本实施例的一些可选的实现方式中，装置600还包括配图单元(附图中未示出)，被配置成：确定文章的主题和实体；获取与文章的主题和实体匹配的图像；根据图像和文章生成图文信息。In some optional implementations of this embodiment, the apparatus 600 further includes an image matching unit (not shown in the drawings), configured to: determine the subject and entity of the article; acquire an image matching the subject and entity of the article ; Generate graphic information from images and articles.
在本实施例的一些可选的实现方式中，装置600还包括排版单元(附图中未示出)，被配置成：将图文信息进行排版优化。In some optional implementations of this embodiment, the apparatus 600 further includes a typesetting unit (not shown in the drawings), which is configured to: perform typesetting optimization on the graphic and text information.
下面参考图7，其示出了适于用来实现本公开的实施例的电子设备(例如图1中的服务器)700的结构示意图。图7示出的服务器仅仅是一个示例，不应对本公开的实施例的功能和使用范围带来任何限制。Referring next to FIG. 7 , it shows a schematic structural diagram of an electronic device (eg, the server in FIG. 1 ) 700 suitable for implementing embodiments of the present disclosure. The server shown in FIG. 7 is only an example, and should not impose any limitation on the function and scope of use of the embodiments of the present disclosure.
如图7所示，电子设备700可以包括处理装置(例如中央处理器、图形处理器等)701，其可以根据存储在只读存储器(ROM)702中的程序或者从存储装置708加载到随机访问存储器(RAM)703中的程序而执行各种适当的动作和处理。在RAM 703中，还存储有电子设备700操作所需的各种程序和数据。处理装置701、ROM 702以及RAM 703通过总线704彼此相连。输入/输出(I/O)接口705也连接至总线704。As shown in FIG. 7 , an electronic device 700 may include a processing device (eg, a central processing unit, a graphics processor, etc.) 701 that may be loaded into random access according to a program stored in a read only memory (ROM) 702 or from a storage device 708 Various appropriate actions and processes are executed by the programs in the memory (RAM) 703 . In the RAM 703, various programs and data necessary for the operation of the electronic device 700 are also stored. The processing device 701 , the ROM 702 , and the RAM 703 are connected to each other through a bus 704 . An input/output (I/O) interface 705 is also connected to bus 704 .
通常，以下装置可以连接至I/O接口705：包括例如触摸屏、触摸板、键盘、鼠标、摄像头、麦克风、加速度计、陀螺仪等的输入装置706；包括例如液晶显示器(LCD)、扬声器、振动器等的输出装置707；包括例如磁带、硬盘等的存储装置708；以及通信装置709。通信装置709可以允许电子设备700与其他设备进行无线或有线通信以交换数据。虽然图7示出了具有各种装置的电子设备700，但是应理解的是，并不要求实施或具备所有示出的装置。可以替代地实施或具备更多或更少的装置。图7中示出的每个方框可以代表一个装置，也可以根据需要代表多个装置。Typically, the following devices can be connected to the I/O interface 705: input devices 706 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; including, for example, a liquid crystal display (LCD), speakers, vibration An output device 707 of a computer, etc.; a storage device 708 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 709. Communication means 709 may allow electronic device 700 to communicate wirelessly or by wire with other devices to exchange data. Although FIG. 7 shows an electronic device 700 having various means, it should be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in FIG. 7 can represent one device, and can also represent multiple devices as required.
特别地，根据本公开的实施例，上文参考流程图描述的过程可以被实现为计算机软件程序。例如，本公开的实施例包括一种计算机程序产品，其包括承载在计算机可读介质上的计算机程序，该计算机程序包含用于执行流程图所示的方法的程序代码。在这样的实施例中，该计算机程序可以通过通信装置709从网络上被下载和安装，或者从存储装置708被安装，或者从ROM 702被安装。在该计算机程序被处理装置701执行时，执行本公开的实施例的方法中限定的上述功能。需要说明的是，本公开的实施例所述的计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质或者是上述两者的任意组合。计算机可读存储介质例如可以是——但不限于——电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子可以包括但不限于：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本公开的实施例中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。而在本公开的实施例中，计算机可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读信号介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括但不限于：电线、光缆、RF(射频)等等，或者上述的任意合适的组合。In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the method illustrated in the flowchart. In such an embodiment, the computer program may be downloaded and installed from the network via the communication device 709 , or from the storage device 708 , or from the ROM 702 . When the computer program is executed by the processing device 701, the above-described functions defined in the methods of the embodiments of the present disclosure are executed. It should be noted that the computer-readable medium described in the embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples of computer readable storage media may include, but are not limited to, electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Programmable read only memory (EPROM or flash memory), fiber optics, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. In embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. Rather, in embodiments of the present disclosure, a computer-readable signal medium may include a data signal in baseband or propagated as part of a carrier wave, carrying computer-readable program code therein. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device . Program code embodied on a computer readable medium may be transmitted using any suitable medium including, but not limited to, electrical wire, optical fiber cable, RF (radio frequency), etc., or any suitable combination of the foregoing.
上述计算机可读介质可以是上述电子设备中所包含的；也可以是单独存在，而未装配入该电子设备中。上述计算机可读介质承载有一个或者多个程序，当上述一个或者多个程序被该电子设备执行时，使得该电子设备：获取待转换的音频信息；将音频信息转换成文本信息；将文本信息进行切词，得到词序列；对于词序列中的词，通过预先训练的词连接概率模型得到的词连接概率表查询该词和与该词的下一个词之间连接概率和该词与各类标点的连接概率，以及基于查询到的连接概率确定该词的连接目标；将词序列中各词与相应的连接目标连接起来生成带标点的文章进行输出。The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or may exist alone without being assembled into the electronic device. The above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by the electronic device, the electronic device: acquires the audio information to be converted; converts the audio information into text information; converts the text information Perform word segmentation to obtain word sequence; for words in the word sequence, query the word connection probability and the connection probability between the word and the next word of the word through the word connection probability table obtained by the pre-trained word connection probability model. The connection probability of the punctuation is determined, and the connection target of the word is determined based on the queried connection probability; each word in the word sequence is connected with the corresponding connection target to generate an article with punctuation for output.
可以以一种或多种程序设计语言或其组合来编写用于执行本公开的实施例的操作的计算机程序代码，所述程序设计语言包括面向对象的程序设计语言—诸如Java、Smalltalk、C++，还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络——包括局域网(LAN)或广域网(WAN)—连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。Computer program code for carrying out operations of embodiments of the present disclosure may be written in one or more programming languages, including object-oriented programming languages—such as Java, Smalltalk, C++, or a combination thereof, Also included are conventional procedural programming languages - such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider through Internet connection).
附图中的流程图和框图，图示了按照本公开各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，该模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more logical functions for implementing the specified functions executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented in dedicated hardware-based systems that perform the specified functions or operations , or can be implemented in a combination of dedicated hardware and computer instructions.
描述于本公开的实施例中所涉及到的单元可以通过软件的方式实现，也可以通过硬件的方式来实现。所描述的单元也可以设置在处理器中，例如，可以描述为：一种处理器包括获取单元、转换单元、切词单元、判断单元和连接单元。其中，这些单元的名称在某种情况下并不构成对该单元本身的限定，例如，获取单元还可以被描述为“获取待转换的音频信息的单元”。The units involved in the embodiments of the present disclosure may be implemented in software or hardware. The described unit can also be set in the processor, for example, it can be described as: a processor includes an acquisition unit, a conversion unit, a word segmentation unit, a judgment unit and a connection unit. Wherein, the names of these units do not constitute a limitation on the unit itself under certain circumstances, for example, the acquisition unit may also be described as "a unit for acquiring audio information to be converted".
以上描述仅为本公开的较佳实施例以及对所运用技术原理的说明。本领域技术人员应当理解，本公开中所涉及的发明范围，并不限于上述技术特征的特定组合而成的技术方案，同时也应涵盖在不脱离所述发明构思的情况下，由上述技术特征或其等同特征进行任意组合而形成的其它技术方案。例如上述特征与本公开中公开的(但不限于)具有类似功能的技术特征进行互相替换而形成的技术方案。The above description is merely a preferred embodiment of the present disclosure and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the present disclosure is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, and should also cover the above-mentioned technical features without departing from the inventive concept. Other technical solutions formed by any combination of its equivalent features. For example, a technical solution is formed by replacing the above features with the technical features disclosed in the present disclosure (but not limited to) with similar functions.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910552619.6ACN110245334B (en) | 2019-06-25 | 2019-06-25 | Method and device for outputting information |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910552619.6ACN110245334B (en) | 2019-06-25 | 2019-06-25 | Method and device for outputting information |
| Publication Number | Publication Date |
|---|---|
| CN110245334Atrue CN110245334A (en) | 2019-09-17 |
| CN110245334B CN110245334B (en) | 2023-06-16 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910552619.6AActiveCN110245334B (en) | 2019-06-25 | 2019-06-25 | Method and device for outputting information |
| Country | Link |
|---|---|
| CN (1) | CN110245334B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111078831A (en)* | 2019-11-06 | 2020-04-28 | 广州荔支网络技术有限公司 | Optimization method for converting audio content into text in text reading |
| CN113297824A (en)* | 2021-05-11 | 2021-08-24 | 北京字跳网络技术有限公司 | Text display method and device, electronic equipment and storage medium |
| WO2022143058A1 (en)* | 2021-01-04 | 2022-07-07 | 北京有竹居网络技术有限公司 | Voice recognition method and apparatus, storage medium, and electronic device |
| CN117093738A (en)* | 2022-11-11 | 2023-11-21 | 上海积图科技有限公司 | Method, system, equipment and storage medium for generating article map distribution material |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070203887A1 (en)* | 2004-06-30 | 2007-08-30 | Misha Dynin | Methods and systems for endorsing search results |
| US20130132401A1 (en)* | 2011-11-17 | 2013-05-23 | Yahoo! Inc. | Related news articles |
| CN106951413A (en)* | 2017-03-24 | 2017-07-14 | 北京百度网讯科技有限公司 | Segmenting method and device based on artificial intelligence |
| CN107221330A (en)* | 2017-05-26 | 2017-09-29 | 北京搜狗科技发展有限公司 | Punctuate adding method and device, the device added for punctuate |
| CN107679039A (en)* | 2017-10-17 | 2018-02-09 | 北京百度网讯科技有限公司 | The method and apparatus being intended to for determining sentence |
| CN107767870A (en)* | 2017-09-29 | 2018-03-06 | 百度在线网络技术（北京）有限公司 | Adding method, device and the computer equipment of punctuation mark |
| CN107944032A (en)* | 2017-12-13 | 2018-04-20 | 北京百度网讯科技有限公司 | Method and apparatus for generating information |
| CN108920611A (en)* | 2018-06-28 | 2018-11-30 | 北京百度网讯科技有限公司 | article generation method, device, equipment and storage medium |
| CN108932220A (en)* | 2018-06-29 | 2018-12-04 | 北京百度网讯科技有限公司 | article generation method and device |
| CN108959556A (en)* | 2018-06-29 | 2018-12-07 | 北京百度网讯科技有限公司 | Entity answering method, device and terminal neural network based |
| CN109213997A (en)* | 2018-08-16 | 2019-01-15 | 昆明理工大学 | A kind of Chinese word cutting method based on two-way length memory network model in short-term |
| CN109241330A (en)* | 2018-08-20 | 2019-01-18 | 北京百度网讯科技有限公司 | The method, apparatus, equipment and medium of key phrase in audio for identification |
| CN109657041A (en)* | 2018-12-04 | 2019-04-19 | 南京理工大学 | The problem of based on deep learning automatic generation method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070203887A1 (en)* | 2004-06-30 | 2007-08-30 | Misha Dynin | Methods and systems for endorsing search results |
| US20130132401A1 (en)* | 2011-11-17 | 2013-05-23 | Yahoo! Inc. | Related news articles |
| CN106951413A (en)* | 2017-03-24 | 2017-07-14 | 北京百度网讯科技有限公司 | Segmenting method and device based on artificial intelligence |
| CN107221330A (en)* | 2017-05-26 | 2017-09-29 | 北京搜狗科技发展有限公司 | Punctuate adding method and device, the device added for punctuate |
| CN107767870A (en)* | 2017-09-29 | 2018-03-06 | 百度在线网络技术（北京）有限公司 | Adding method, device and the computer equipment of punctuation mark |
| CN107679039A (en)* | 2017-10-17 | 2018-02-09 | 北京百度网讯科技有限公司 | The method and apparatus being intended to for determining sentence |
| CN107944032A (en)* | 2017-12-13 | 2018-04-20 | 北京百度网讯科技有限公司 | Method and apparatus for generating information |
| CN108920611A (en)* | 2018-06-28 | 2018-11-30 | 北京百度网讯科技有限公司 | article generation method, device, equipment and storage medium |
| CN108932220A (en)* | 2018-06-29 | 2018-12-04 | 北京百度网讯科技有限公司 | article generation method and device |
| CN108959556A (en)* | 2018-06-29 | 2018-12-07 | 北京百度网讯科技有限公司 | Entity answering method, device and terminal neural network based |
| CN109213997A (en)* | 2018-08-16 | 2019-01-15 | 昆明理工大学 | A kind of Chinese word cutting method based on two-way length memory network model in short-term |
| CN109241330A (en)* | 2018-08-20 | 2019-01-18 | 北京百度网讯科技有限公司 | The method, apparatus, equipment and medium of key phrase in audio for identification |
| CN109657041A (en)* | 2018-12-04 | 2019-04-19 | 南京理工大学 | The problem of based on deep learning automatic generation method |
| Title |
|---|
| 侯加英: ""汉越双语新闻话题发现研究"", 《中国优秀硕士学位论文全文数据库信息科技辑》, pages 138 - 1937* |
| 江跃华等: ""融合词汇特征的生成式摘要模型"", 《河北科技大学学报》, pages 152 - 158* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111078831A (en)* | 2019-11-06 | 2020-04-28 | 广州荔支网络技术有限公司 | Optimization method for converting audio content into text in text reading |
| CN111078831B (en)* | 2019-11-06 | 2023-05-30 | 广州荔支网络技术有限公司 | Optimizing method for converting text from text reading audio content |
| WO2022143058A1 (en)* | 2021-01-04 | 2022-07-07 | 北京有竹居网络技术有限公司 | Voice recognition method and apparatus, storage medium, and electronic device |
| CN113297824A (en)* | 2021-05-11 | 2021-08-24 | 北京字跳网络技术有限公司 | Text display method and device, electronic equipment and storage medium |
| CN117093738A (en)* | 2022-11-11 | 2023-11-21 | 上海积图科技有限公司 | Method, system, equipment and storage medium for generating article map distribution material |
| Publication number | Publication date |
|---|---|
| CN110245334B (en) | 2023-06-16 |
| Publication | Publication Date | Title |
|---|---|---|
| CN108022586B (en) | Method and apparatus for controlling pages | |
| CN115082602A (en) | Method for generating digital human, training method, device, equipment and medium of model | |
| CN110245334B (en) | Method and device for outputting information | |
| CN109858045B (en) | Machine translation method and device | |
| CN111382228A (en) | Method and apparatus for outputting information | |
| JP7140913B2 (en) | Video distribution statute of limitations determination method and device | |
| WO2020182123A1 (en) | Method and device for pushing statement | |
| WO2022156434A1 (en) | Method and apparatus for generating text | |
| CN112182255A (en) | Method and apparatus for storing and retrieving media files | |
| CN111026849B (en) | Data processing method and device | |
| CN111582360A (en) | Method, apparatus, device and medium for labeling data | |
| CN110457325A (en) | Method and device for outputting information | |
| CN111008213A (en) | Method and apparatus for generating language conversion model | |
| CN113761865A (en) | Sound and text realignment and information presentation method and device, electronic equipment and storage medium | |
| US20160379636A1 (en) | System and method for handling a spoken user request | |
| WO2024099171A1 (en) | Video generation method and apparatus | |
| CN111639234B (en) | Method and apparatus for mining core entity concerns | |
| CN112148848A (en) | A question and answer processing method and device | |
| CN111555960A (en) | Method for generating information | |
| CN111078849A (en) | Method and apparatus for outputting information | |
| US20210217437A1 (en) | Method and apparatus for processing voice | |
| CN112328896B (en) | Method, apparatus, electronic device, and medium for outputting information | |
| CN110598049A (en) | Method, apparatus, electronic device and computer readable medium for retrieving video | |
| CN112651231B (en) | Spoken language information processing methods, devices and electronic equipment | |
| CN113241061A (en) | Method and device for processing voice recognition result, electronic equipment 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 |