Audio Recorder

The audio recorder is a small JavaScript library that captures audio from the microphone, resamples it at a the desired sample rate, and feeds it to any desired audio application (such as pocketsphinx.js). It makes use of the web audio API, WebRTC and web workers and it is derived from Recorderjs. It works on Chrome and Firefox (25+). To know more about audio capture and playback on the web, you can have a look at this overview of audio on the Web.

1. Files

You need to include both audioRecorder.js and audioRecorderWorker.js in your web application. You can find them in the webapp/js/ folder. The default parameters assume that audioRecorderWorker.js is in a js/ subfolder starting from the HTML file. You can change this with a parameter to pass at initialization time.

2. Usage

In your HTML, include audioRecorder.js:

<script src="js/audioRecorder.js"></script>

When the page is loaded, you can initialize the recorder. To create a new instance of AudioRecorder, you need an input node generated from a stream created by getUserMedia. So, what you need to do is:

• Get an instance of AudioContext.
• Create a callback function to pass to getUserMedia in case of success (and also one in case of failure).
• In the callback, create a web audio API node with createMediaStreamSource using the stream given by getUserMedia.
• In the callback, instantiate the AudioRecorder using the newly created node.
• Attach consumers to the recorder.
• Call getUserMedia, with {audio: true} as parameter.

In practice, it could look like this:

// Deal with prefixed APIs
window.AudioContext = window.AudioContext || window.webkitAudioContext;
navigator.getUserMedia = navigator.getUserMedia ||
                         navigator.webkitGetUserMedia ||
                         navigator.mozGetUserMedia;
    
// Instantiating AudioContext
try {
    var audioContext = new AudioContext();
} catch (e) {
    console.log("Error initializing Web Audio");
}

var recorder;
// Callback once the user authorizes access to the microphone:
function startUserMedia(stream) {
    var input = audioContext.createMediaStreamSource(stream);
    recorder = new AudioRecorder(input);
    // We can, for instance, add a recognizer as consumer
    if (recognizer) recorder.consumers.push(recognizer);
 };
    
// Actually call getUserMedia
if (navigator.getUserMedia)
    navigator.getUserMedia({audio: true},
                           startUserMedia,
                           function(e) {console.log("No live audio input in this browser");}
                          );
else console.log("No web audio support in this browser");

3. Parameters

In the previous section, we kept the default values of the parameters, but a config object can be used at initialization time:

recorder = new AudioRecorder(input, config);

where config can have the following fields:

• errorCallback: Callback when the recorder encounters an issue. The callback takes one argument which is a string that describes the error. For now, the only possible error is the self-explanatory "silent".
• inputBufferLength: Sets the length of buffer to use for audio input (defaults to 4096, no matter the input sample rate).
• outputBufferLength: Sets the length of buffer to send to consumers (defaults to 4000, no matter the output sample rate).
• outputSampleRate: Sets the output sample rate for the data to be sent to the consumers (defaults to 16000Hz). The output sample rate should not be greater than the input sample rate (which depends on the audio hardware, usually 44.1kHz or 48kHz).
• worker: Location of the audioRecorderWorker.js file, starting from the HTML file that loads audioRecorder.js (defaults to js/audioRecorderWorker.js).

4. API

In this section, we assume we have an AudioRecorder instance called recorder.

4.a Start, Stop, Cancel

• To start recording, call recorder.start(data), where data is anything you want to send to the consumers when recording starts. For instance if your consumer is a speech recognizer, you might want to pass the language model to use.
• To stop, use recorder.stop() or recorder.cancel().

4.b Consumers

Audio samples generated by the audio recorder are then passed to consumers. You can have many consumers listening to the data. For instance, if you have a speech recognizer and an audio visualizer:

recorder.consumers = [recognizer, visualizer];

Consumers should be web workers that understand the following messages:

• {command: 'start', data: data}, posted when recorder.start(data) is called. data can be anything useful to the consumer. For instance it can indicate which language model to use to a speech recognizer.
• {command: 'stop'}, posted when recorder.stop() is called.
• {command: 'process', data: audioSamples}, posted every time an audio buffer is available, the buffer is stored in data. The buffer is a Int16Array typed array. If you need to pass it in a message for Google's PPAPI, you might need to take the underlying ArrayBuffer with audioSamples.buffer.

5. License

The files webapp/js/audioRecorder.js and webapp/js/audioRecorderWorker.js are based on Recorder.js, which is under the MIT license (Copyright © 2013 Matt Diamond).

The remaining of this software is licensed under the MIT license:

Copyright © 2013-2014 Sylvain Chevalier

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.