MozillaWiki

User:David.humphrey/Audio Data API 2: Difference between revisions

User page Discussion

User:David.humphrey/Audio Data API 2 (view source)

Revision as of 01:32, 23 May 2010

5,330 bytes added ,  23 May 2010
→‎Reading Audio
 
(18 intermediate revisions by the same user not shown)
Line 33: Line 33:
* Native array interfaces instead of using accessors and IDL array arguments.
* Native array interfaces instead of using accessors and IDL array arguments.
* No zero padding of audio data occurs anymore.  All frames are exactly 4096 elements in length.
* No zero padding of audio data occurs anymore.  All frames are exactly 4096 elements in length.
* Added '''mozSampleOffset()'''
* Added '''mozCurrentSampleOffset()'''
* Removed undocumented position/buffer methods on audio element.
* Removed undocumented position/buffer methods on audio element.
* Added '''mozChannels''' and '''mozRate''' to '''loadedmetadata' event.
* Added '''mozChannels''', '''mozRate''', '''mozFrameBufferLength''' to '''loadedmetadata' event.


Demos written for the previous version are '''not''' compatible, though can be made to be quite easily.  See details below.
Demos written for the previous version are '''not''' compatible, though can be made to be quite easily.  See details below.
Line 65: Line 65:
var channels,
var channels,
     rate,
     rate,
    frameBufferLength,
     samples;
     samples;


function audioInfo(event) {
function audioInfo(event) {
   channels = event.mozChannels;
   channels         = event.mozChannels;
   rate     = event.mozRate;
   rate             = event.mozRate;
  frameBufferLength = event.mozFrameBufferLength;
}
}


Line 85: Line 87:
===== Complete Example: Visualizing Audio Spectrum =====
===== Complete Example: Visualizing Audio Spectrum =====


'''NOTE: This example is not updated to the new API.'''
This example calculates and displays FFT spectrum data for the playing audio:
 
This example uses the native FFT data from <code>mozSpectrum</code> to display the frequency spectrum in a canvas:


[[File:fft.png]]
[[File:fft.png]]
Line 94: Line 94:
<!DOCTYPE html>
<!DOCTYPE html>
<html>
<html>
   <head>    
   <head>
     <title>JavaScript Spectrum Example</title>
     <title>JavaScript Spectrum Example</title>
   </head>
   </head>
   <body>
   <body>
     <audio src="song.ogg"  
     <audio src="song.ogg"
           controls="true"
           controls="true"
          onloadedmetadata="loadedMetadata(event);"
           onaudiowritten="audioWritten(event);"
           onaudiowritten="audioWritten(event);"
           style="width: 512px;">
           style="width: 512px;">
     </audio>
     </audio>
   
     <div><canvas id="fft" width="512" height="200"></canvas></div>
     <div><canvas id="fft" width="512" height="200"></canvas></div>
   
 
     <script>
     <script>
      var spectrum;
       var canvas = document.getElementById('fft'),
     
          ctx = canvas.getContext('2d'),
       var canvas = document.getElementById('fft');
          fft;
      var ctx = canvas.getContext('2d');
 
      function loadedMetadata(event) {
        var channels          = event.mozChannels,
            rate              = event.mozRate,
            frameBufferLength = event.mozFrameBufferLength;
       
        fft = new FFT(frameBufferLength / channels, rate),
      }


       function audioWritten(event) {
       function audioWritten(event) {
         spectrum = event.mozSpectrum;
         var fb = event.mozFrameBuffer,
       
            signal = new Float32Array(fb.length / channels),
         var specSize = spectrum.length, magnitude;
            magnitude;
          
 
         for (var i = 0, fbl = fb.length / 2; i < fbl; i++ ) {
          // Assuming interlaced stereo channels,
          // need to split and merge into a stero-mix mono signal
          signal[i] = (fb[2*i] + fb[2*i+1]) / 2;
         }
 
        fft.forward(signal);
 
         // Clear the canvas before drawing spectrum
         // Clear the canvas before drawing spectrum
         ctx.clearRect(0,0, canvas.width, canvas.height);
         ctx.clearRect(0,0, canvas.width, canvas.height);
       
 
         for ( var i = 0; i < specSize; i++ ) {
         for (var i = 0; i < fft.spectrum.length; i++ ) {
           magnitude = spectrum.item(i) * 4000; // multiply spectrum by a zoom value
           // multiply spectrum by a zoom value
            
           magnitude = fft.spectrum[i] * 4000;
 
           // Draw rectangle bars for each frequency bin
           // Draw rectangle bars for each frequency bin
           ctx.fillRect(i * 4, canvas.height, 3, -magnitude);
           ctx.fillRect(i * 4, canvas.height, 3, -magnitude);
         }
         }
       }
       }
      // FFT from dsp.js, see below
      var FFT = function(bufferSize, sampleRate) {
        this.bufferSize  = bufferSize;
        this.sampleRate  = sampleRate;
        this.spectrum    = new Float32Array(bufferSize/2);
        this.real        = new Float32Array(bufferSize);
        this.imag        = new Float32Array(bufferSize);
        this.reverseTable = new Uint32Array(bufferSize);
        this.sinTable    = new Float32Array(bufferSize);
        this.cosTable    = new Float32Array(bufferSize);
        var limit = 1,
            bit = bufferSize >> 1;
        while ( limit < bufferSize ) {
          for ( var i = 0; i < limit; i++ ) {
            this.reverseTable[i + limit] = this.reverseTable[i] + bit;
          }
          limit = limit << 1;
          bit = bit >> 1;
        }
        for ( var i = 0; i < bufferSize; i++ ) {
          this.sinTable[i] = Math.sin(-Math.PI/i);
          this.cosTable[i] = Math.cos(-Math.PI/i);
        }
      };
      FFT.prototype.forward = function(buffer) {
        var bufferSize  = this.bufferSize,
            cosTable    = this.cosTable,
            sinTable    = this.sinTable,
            reverseTable = this.reverseTable,
            real        = this.real,
            imag        = this.imag,
            spectrum    = this.spectrum;
        if ( bufferSize !== buffer.length ) {
          throw "Supplied buffer is not the same size as defined FFT. FFT Size: " +
                bufferSize + " Buffer Size: " + buffer.length;
        }
        for ( var i = 0; i < bufferSize; i++ ) {
          real[i] = buffer[reverseTable[i]];
          imag[i] = 0;
        }
        var halfSize = 1,
            phaseShiftStepReal,
            phaseShiftStepImag,
            currentPhaseShiftReal,
            currentPhaseShiftImag,
            off,
            tr,
            ti,
            tmpReal,
            i;
        while ( halfSize < bufferSize ) {
          phaseShiftStepReal = cosTable[halfSize];
          phaseShiftStepImag = sinTable[halfSize];
          currentPhaseShiftReal = 1.0;
          currentPhaseShiftImag = 0.0;
          for ( var fftStep = 0; fftStep < halfSize; fftStep++ ) {
            i = fftStep;
            while ( i < bufferSize ) {
              off = i + halfSize;
              tr = (currentPhaseShiftReal * real[off]) - (currentPhaseShiftImag * imag[off]);
              ti = (currentPhaseShiftReal * imag[off]) + (currentPhaseShiftImag * real[off]);
              real[off] = real[i] - tr;
              imag[off] = imag[i] - ti;
              real[i] += tr;
              imag[i] += ti;
              i += halfSize << 1;
            }
            tmpReal = currentPhaseShiftReal;
            currentPhaseShiftReal = (tmpReal * phaseShiftStepReal) - (currentPhaseShiftImag * phaseShiftStepImag);
            currentPhaseShiftImag = (tmpReal * phaseShiftStepImag) + (currentPhaseShiftImag * phaseShiftStepReal);
          }
          halfSize = halfSize << 1;
}
        i = bufferSize/2;
        while(i--) {
          spectrum[i] = 2 * Math.sqrt(real[i] * real[i] + imag[i] * imag[i]) / bufferSize;
}
      };
     </script>
     </script>
   </body>
   </body>
Line 157: Line 268:
</pre>
</pre>


<code>mozSampleOffset()</code>
<code>mozCurrentSampleOffset()</code>


<pre>
<pre>
// Get current position of the underlying audio stream, measured in samples written.
// Get current position of the underlying audio stream, measured in samples written.
var currentSampleOffset = audioOutput.mozSampleOffset();
var currentSampleOffset = audioOutput.mozCurrentSampleOffset();
</pre>
</pre>


Line 167: Line 278:


<pre>
<pre>
// Create a new audio stream for writing
<audio id="a1"
var aout = new Audio();
      src="song.ogg"
aout.mozSetup(2, 44100, 1);
      onloadedmetadata="loadedMetadata(event);"
      onaudiowritten="audioWritten(event);"
      controls="controls">
</audio>
<script>
var a1 = document.getElementById('a1'),
    a2 = new Audio(),


function audioWritten(event){
function loadedMetadata(event) {
   samples = event.mozFrameBuffer;
   // Mute a1 audio.
  a1.volume = 0;
  // Setup a2 to be identical to a1, and play through there.
  a2.mozSetup(event.mozChannels, event.mozRate, 1);
}


  // Do any filtering, signal processing, etc.
function audioWritten(event) {
  for (var i=0, slen=samples.length; i < slen; i++){
   // Write the current frame to a2
    process(samples[i]);
   a2.mozWriteAudio(event.mozFrameBuffer);
   } 
 
   aout.mozWriteAudio(samples);
}
}
</script>
</pre>
</pre>
Audio data written using the '''mozWriteAudio()''' method needs to be written at a regular interval in equal portions, in order to keep a little ahead of the current sample offset (current sample offset of hardware can be obtained with '''mozCurrentSampleOffset()'''), where a little means something on the order of 500ms of samples.  For example, if working with 2 channels at 44100 samples per second, and a writing interval chosen that is equal to 100ms, and a pre-buffer equal to 500ms, one would write an array of (2 * 44100 / 10) = 8820 samples, and a total of (currentSampleOffset + 2 * 44100 / 2).


===== Complete Example: Creating a Web Based Tone Generator =====
===== Complete Example: Creating a Web Based Tone Generator =====
Line 195: Line 316:
   <body>
   <body>
     <input type="text" size="4" id="freq" value="440"><label for="hz">Hz</label>
     <input type="text" size="4" id="freq" value="440"><label for="hz">Hz</label>
    <button onclick="generateWaveform()">set</button>
     <button onclick="start()">play</button>
     <button onclick="start()">play</button>
     <button onclick="stop()">stop</button>
     <button onclick="stop()">stop</button>


     <script type="text/javascript">
     <script type="text/javascript">
       var sampledata = [];
       var sampleRate = 44100,
      var freq = 440;
          portionSize = sampleRate / 10,
      var interval = -1;
          prebufferSize = sampleRate / 2,
      var audio;
          freq = undefined; // no sound


       function writeData() {
       var audio = new Audio();
        var n = Math.ceil(freq / 100);
      audio.mozSetup(1, sampleRate, 1);
        for(var i=0;i<n;i++)
       var currentWritePosition = 0;
          audio.mozWriteAudio(sampledata);
       }


       function start() {
       function getSoundData(t, size) {
         audio = new Audio();
         var soundData = new Float32Array(size);
         audio.mozSetup(1, 44100, 1);
         if (freq) {
        interval = setInterval(writeData, 10);
          var k = 2* Math.PI * freq / sampleRate;
          for (var i=0; i<size; i++) {
            soundData[i] = Math.sin(k * (i + t));
          }
        }
        return soundData;
       }
       }


       function stop() {
       function writeData() {
         if (interval != -1) {
         while(audio.mozCurrentSampleOffset() + prebufferSize >= currentWritePosition) {
           clearInterval(interval);
           var soundData = getSoundData(currentWritePosition, portionSize);
           interval = -1;
          audio.mozWriteAudio(soundData);
           currentWritePosition += portionSize;
         }
         }
       }
       }


       function generateWaveform() {
      // initial write
      writeData();
      var writeInterval = Math.floor(1000 * portionSize / sampleRate);
      setInterval(writeData, writeInterval);
 
       function start() {
         freq = parseFloat(document.getElementById("freq").value);
         freq = parseFloat(document.getElementById("freq").value);
        // we're playing at 44.1kHz, so figure out how many samples
        // will give us one full period
        var samples = 44100 / freq;
        sampledata = Array(Math.round(samples));
        for (var i=0; i<sampledata.length; i++) {
          sampledata[i] = Math.sin(2*Math.PI * (i / sampledata.length));
        }
       }
       }


       generateWaveform();
       function stop() {
        freq = undefined;
      }
   </script>
   </script>
   </body>
   </body>
Line 257: Line 381:
   readonly attribute unsigned long mozChannels;
   readonly attribute unsigned long mozChannels;
   readonly attribute unsigned long mozRate;
   readonly attribute unsigned long mozRate;
  readonly attribute unsigned long mozFrameBufferLength;
};
};
</pre>
</pre>


The '''mozChannels''' attribute contains a the number of channels in this audio resource (e.g., 2).  The '''mozRate''' attribute contains the number of samples per second that will be played, for example 44100.
The '''mozChannels''' attribute contains a the number of channels in this audio resource (e.g., 2).  The '''mozRate''' attribute contains the number of samples per second that will be played, for example 44100.  The '''mozFrameBufferLength''' attribute contains the number of samples that will be returned in each '''AudioWritten''' event.  This number is a total for all channels (e.g., 2 channels * 2048 samples = 4096 total).


===== nsIDOMNotifyAudioWrittenEvent =====
===== nsIDOMNotifyAudioWrittenEvent =====
Line 335: Line 460:
=== JavaScript Audio Libraries ===
=== JavaScript Audio Libraries ===


We have started work on a JavaScript library to make building audio web apps easier.  Details are [[Audio Data API JS Library|here]].
* We have started work on a JavaScript library to make building audio web apps easier.  Details are [[Audio Data API JS Library|here]].
* [http://github.com/bfirsh/dynamicaudio.js dynamicaudio.js] - An interface for writing audio with a Flash fall back for older browsers.


=== Working Audio Data Demos ===
=== Working Audio Data Demos ===
Line 347: Line 473:
* FFT visualization (calculated with js)
* FFT visualization (calculated with js)
** http://weare.buildingsky.net/processing/dsp.js/examples/fft.html
** http://weare.buildingsky.net/processing/dsp.js/examples/fft.html
* Beat Detection (also showing use of WebGL for 3D visualizations)
** http://cubicvr.org/CubicVR.js/bd3/BeatDetektor1HD-13a.html (video [http://vimeo.com/11345262 here])
** http://cubicvr.org/CubicVR.js/bd3/BeatDetektor2HD-13a.html (video of older version [http://vimeo.com/11345685 here])
** http://cubicvr.org/CubicVR.js/bd3/BeatDetektor3HD-13a.html (video [http://www.youtube.com/watch?v=OxoFcyKYwr0&fmt=22 here])


* Writing Audio from JavaScript, Digital Signal Processing
* Writing Audio from JavaScript, Digital Signal Processing
** Csound shaker instrument ported to JavaScript via Processing.js http://scotland.proximity.on.ca/dxr/tmp/audio/shaker/
** Csound shaker instrument ported to JavaScript via Processing.js http://scotland.proximity.on.ca/dxr/tmp/audio/shaker/


==== Demos Needing to be Updated to New API ====
==== Demos Needing to be Updated to New API ====
Line 362: Line 492:
** http://ondras.zarovi.cz/demos/audio/
** http://ondras.zarovi.cz/demos/audio/


* Beat Detection (also showing use of WebGL for 3D visualizations)
** http://cubicvr.org/CubicVR.js/BeatDetektor1HD.html (video [http://vimeo.com/11345262 here])
** http://cubicvr.org/CubicVR.js/BeatDetektor2HD.html (video [http://vimeo.com/11345685 here])
** http://weare.buildingsky.net/processing/beat_detektor/beat_detektor.html
** http://weare.buildingsky.net/processing/beat_detektor/beat_detektor.html
** http://code.bocoup.com/processing-js/3d-fft/viz.xhtml
** http://code.bocoup.com/processing-js/3d-fft/viz.xhtml
David.humphrey
Confirmed users
656

edits

Retrieved from "https://wiki.allizom.org/Special:MobileDiff/224776...225059"