- Notifications
You must be signed in to change notification settings - Fork19
A web audio sampler instrument
danigb/smplr
Folders and files
Name | Name | Last commit message | Last commit date | |
---|---|---|---|---|
Repository files navigation
smplr
is a collection of sampled instruments for Web Audio API ready to be used with no setup required.
Examples:
import{Soundfont}from"smplr";constcontext=newAudioContext();constmarimba=newSoundfont(context,{instrument:"marimba"});marimba.start({note:60,velocity:80});
import{DrumMachine}from"smplr";constcontext=newAudioContext();constdm=newDrumMachine(context);dm.start({note:"kick"});
import{SplendidGrandPiano,Reverb}from"smplr";constcontext=newAudioContext();constpiano=newSplendidGrandPiano(context);piano.output.addEffect("reverb",newReverb(context),0.2);piano.start({note:"C4"});
See demo:https://danigb.github.io/smplr/
smplr
is still under development and features are considered unstable until v 1.0
ReadCHANGELOG for changes.
- No setup: specifically, all samples are online, so no need for a server.
- Easy to use: everything should be intuitive for non-experienced developers
- Decent sounding: uses high quality open source samples. For better or worse, it is sample based 🤷
You can install the library with a package manager or use it directly by importing from the browser.
Samples are stored athttps://github.com/smpldsnds and there is no need to download them. Kudos to allsamplerist 🙌
Use npm or your favourite package manager to install the library to use it in your project:
npm i smplr
You can import directly from the browser. For example:
<html><body><buttonid="btn">play</button></body><scripttype="module">import{SplendidGrandPiano}from"https://unpkg.com/smplr/dist/index.mjs";// needs to be a urlconstcontext=newAudioContext();// create the audio contextconstmarimba=newSplendidGrandPiano(context);// create and load the instrumentdocument.getElementById("btn").onclick=()=>{context.resume();// enable audio context after a user interactionmarimba.start({note:60,velocity:80});// play the note};</script></html>
The package needs to be serve as a url from a service likeunpkg or similar.
All instruments follows the same pattern:new Instrument(context, options)
. For example:
import{SplendidGrandPiano,Soundfont}from"smplr";constcontext=newAudioContext();constpiano=newSplendidGrandPiano(context,{decayTime:0.5});constmarimba=newSoundfont(context,{instrument:"marimba"});
You can start playing notes as soon as one audio is loaded. But if you want to wait for all of them, you can use theload
property that returns a promise:
piano.load.then(()=>{// now the piano is fully loaded});
Since the promise returns the instrument instance, you can create and wait in a single line:
constpiano=awaitnewSplendidGrandPiano(context).load;
loaded()
function was exposed instead.
All instruments share some configuration options that are passed as second argument of the constructor. As it name implies, all fields are optional:
volume
: A number from 0 to 127 representing the instrument global volume. 100 by defaultdestination
: AnAudioNode
that is the output of the instrument.AudioContext.destination
is used by defaultvolumeToGain
: a function to convert the volume to gain. It uses MIDI standard as default.disableScheduler
: disable internal scheduler.false
by default.scheduleLookaheadMs
: the lookahead of the scheduler. If the start time of the note is less than current time plus this lookahead time, the note will be started. 200ms by default.scheduleIntervalMs
: the interval of the scheduler. 50ms by default.onStart
: a function that is called when starting a note. It receives the note started as parameter. Bear in mind that the time this function is called is not precise, and it's determined by lookahead.onEnded
: a function that is called when the note ends. It receives the started note as parameter.
This package should be compatible withstandardized-audio-context:
import{AudioContext}from"standardized-audio-context";constcontext=newAudioContext();constpiano=newSplendidGrandPiano(context);
However, if you are using Typescript, you might need to "force cast" the types:
import{Soundfont}from"smplr";import{AudioContextasStandardizedAudioContext}from"standardized-audio-context";constcontext=newStandardizedAudioContext()asunknownasAudioContext;constmarimba=newSoundfont(context,{instrument:"marimba"});
Thestart
function accepts a bunch of options:
piano.start({note:"C4",velocity:80,time:5,duration:1});
Thevelocity
is a number between 0 and 127 the represents at which velocity the key is pressed. The bigger the number, louder the sound. Butvelocity
not only controls the loudness. In some instruments, it also affects the timbre.
Thestart
function returns astop
function for the given note:
conststopNote=piano.start({note:60});stopNote({time:10});
Bear in mind that you may need to callcontext.resume()
before playing a note
Instruments have a globalstop
function that can be used to stop all notes:
// This will stop all notespiano.stop();
Or stop the specified one:
// This will stop C4 notepiano.stop(60);
You can schedule notes usingtime
andduration
properties. Both are measured in seconds. Time is the number of seconds since the AudioContext was created, like inaudioContext.currentTime
For example, next example plays a C major arpeggio, one note per second:
constnow=context.currentTime;["C4","E4","G4","C5"].forEach((note,i)=>{piano.start({ note,time:now+i,duration:0.5});});
You can loop a note by usingloop
,loopStart
andloopEnd
:
constsampler=newSampler(audioContext,{duh:"duh-duh-ah.mp3"});sampler.start({note:"duh"loop:trueloopStart:1.0,loopEnd:9.0,});
Ifloop
is true butloopStart
orloopEnd
are not specified, 0 and total duration will be used by default, respectively.
Instrumentoutput
attribute represents the main output of the instrument.output.setVolume
method accepts a number where 0 means no volume, and 127 is max volume without amplification:
piano.output.setVolume(80);
volume
is global to the instrument, butvelocity
is specific for each note.
Two events are supportedonStart
andonEnded
. Both callbacks will receive as parameter started note.
Events can be configured globally:
constcontext=newAudioContext();constsampler=newSample(context,{onStart:(note)=>{console.log(note.time,context.currentTime);},});
or per note basis:
piano.start({note:"C4",duration:1,onEnded:()=>{// will be called after 1 second},});
Global callbacks will be invoked regardless of whether local events are defined.
onStart
is not exact. It triggers slightly before the actual start time and is influenced by thescheduleLookaheadMs
parameter.
An packed version ofDattorroReverbNode algorithmic reverb is included.
Useoutput.addEffect(name, effect, mix)
to connect an effect using a send bus:
import{Reverb,SplendidGrandPiano}from"smplr";constreverb=newReverb(context);constpiano=newSplendidGrandPiano(context,{ volume});piano.output.addEffect("reverb",reverb,0.2);
To change the mix level, useoutput.sendEffect(name, mix)
:
piano.output.sendEffect("reverb",0.5);
If you use default samples, they are stored at github pages. Github rate limits the number of requests per second. That could be a problem, specially if you're using a development environment with hot reload (like most React frameworks).
If you want to cache samples on the browser you can use aCacheStorage
object:
import{SplendidGrandPiano,CacheStorage}from"smplr";constcontext=newAudioContext();conststorage=newCacheStorage();// First time the instrument loads, will fetch the samples from http. Subsequent times from cache.constpiano=newSplendidGrandPiano(context,{ storage});
CacheStorage
is based onCache API and only works in secure environments that runs withhttps
. Read your framework documentation for setup instructions. For example, in nextjs you can usehttps://www.npmjs.com/package/next-dev-https. For vite there'shttps://github.com/liuweiGL/vite-plugin-mkcert. Find the appropriate solution for your environment.
An audio buffer sampler. Pass abuffers
object with the files to be load:
import{Sampler}from"smplr";constbuffers={kick:"https://smpldsnds.github.io/drum-machines/808-mini/kick.m4a",snare:"https://smpldsnds.github.io/drum-machines/808-mini/snare-1.m4a",};constsampler=newSampler(newAudioContext(),{ buffers});
And then use the name of the buffer as note name:
sampler.start({note:"kick"});
A Soundfont player. By default it loads audio from Benjamin Gleitzman's package ofpre-rendered sound fonts.
import{Soundfont,getSoundfontNames,getSoundfontKits}from"smplr";constmarimba=newSoundfont(newAudioContext(),{instrument:"marimba"});marimba.start({note:"C4"});
It's intended to be a modern replacement ofsoundfont-player
UsegetSoundfontNames
to get all available instrument names andgetSoundfontKits
to get kit names.
There are two kits available:MusyngKite
orFluidR3_GM
. The first one is used by default: it sounds better but samples weights more.
constmarimba=newSoundfont(context,{instrument:"clavinet",kit:"FluidR3_GM",// "MusyngKite" is used by default if not specified});
Alternatively, you can pass your custom url as the instrument. In that case, thekit
is ignored:
constmarimba=newSoundfont(context,{instrumentUrl:"https://gleitz.github.io/midi-js-soundfonts/MusyngKite/marimba-mp3.js",});
You can enable note looping to make note names indefinitely long by loading loop data:
constmarimba=newSoundfont(context,{instrument:"cello",loadLoopData:true,});
A sampled acoustic piano. It uses Steinway samples with 4 velocity groups fromSplendidGrandPiano
import{SplendidGrandPiano}from"smplr";constpiano=newSplendidGrandPiano(newAudioContext());piano.start({note:"C4"});
The second argument of the constructor accepts the following options:
baseUrl
:detune
: global detune in cents (0 if not specified)velocity
: default velocity (100 if not specified)volume
: default volume (100 if not specified)decayTime
: default decay time (0.5 seconds)notesToLoad
: an object with the following shape:{ notes: number[], velocityRange: [number, number]}
to specify a subset of notes to load
Example:
constpiano=newSplendidGrandPiano(context,{detune:-20,volume:80,notesToLoad:{notes:[60],velocityRange:[1,127],},});
A sampled electric pianos. Samples fromhttps://github.com/sfzinstruments/GregSullivan.E-Pianos
import{ElectricPiano,getElectricPianoNames}from"smplr";constinstruments=getElectricPianoNames();// => ["CP80", "PianetT", "WurlitzerEP200"]constepiano=newElectricPiano(newAudioContext(),{instrument:"PianetT",});epiano.start({note:"C4"});// Includes a (basic) tremolo effect:epiano.tremolo.level(30);
Available instruments:
CP80
: Yamaha CP80 Electric Grand Piano v1.3 (29-Sep-2004)PianetT
: Hohner Pianet T (type 2) v1.3 (24-Sep-2004)WurlitzerEP200
: Wurlitzer EP200 Electric Piano v1.1 (16-May-1999)
Samples fromThe Versilian Community Sample Library
import{Mallet,getMalletNames}from"smplr";constinstruments=getMalletNames();constmallet=newMallet(newAudioContext(),{instrument:instruments[0],});
Samples fromarchive.org
import{Mellotron,getMellotronNames}from"smplr";constinstruments=getMellotronNames();constmallet=newMellotron(newAudioContext(),{instrument:instruments[0],});
Sampled drum machines. Samples from different sources:
import{DrumMachine,getDrumMachineNames}from"smplr";constinstruments=getDrumMachineNames();constcontext=newAudioContext();constdrums=newDrumMachine(context,{instrument:"TR-808"});drums.start({note:"kick"});// Drum samples are grouped and can have sample variations:drums.getSampleNames();// => ['kick-1', 'kick-2', 'snare-1', 'snare-2', ...]drums.getGroupNames();// => ['kick', 'snare']drums.getSampleNamesForGroup("kick")=>// => ['kick-1', 'kick-2']// You can trigger samples by group name or specific sampledrums.start("kick");// Play the first sample of the groupdrums.start("kick-1");// Play this specific sample
import{Smolken,getSmolkenNames}from"smplr";constinstruments=getSmolkenNames();// => Arco, Pizzicato & Switched// Create an instrumentconstcontext=newAudioContext();constdoubleBass=awaitnewSmolken(context,{instrument:"Arco"}).load;
Versilian is a sample capable of using theVersilian Community Sample Library.
import{Versilian,getVersilianInstruments}from"smplr";// getVersilianInstruments returns a PromiseconstinstrumentNames=awaitgetVersilianInstruments();constcontext=newAudioContext();constsampler=newVersilian(context,{instrument:instrumentNames[0]});
Sampler capable of reading .sf2 files directly:
import{Soundfont2Sampler}from"smplr";import{SoundFont2}from"soundfont2";constcontext=newAudioContext();constsampler=newSoundfont2Sampler(context,{url:"https://smpldsnds.github.io/soundfonts/soundfonts/galaxy-electric-pianos.sf2",createSoundfont:(data)=>newSoundFont2(data),});sampler.load.then(()=>{// list all available instruments for the soundfontconsole.log(sampler.instrumentNames);// load the first available instrumentsampler.loadInstrument(sampler.instrumentNames[0]);});
Still limited support. API may vary.
MIT License
About
A web audio sampler instrument