Getting the volume level
yydad117 opened this issue · comments
Hi, I like your plugin and it is possible to get the volume level directly from the microphone? Or detect the volume change with the plugins?
It of course depends how accurate you need it to be, but you can get the current volume level registered by the microphone from the plugin.
One way of doing it is to use the Web Audio (AudioNode) based method that the plugin supports. The following code should be read as psuedo-code since I just wrote it here in this issue and have not actually run it live. It can contain some errors:
// Start the plugin in Web Audio mode
audioinput.start({
streamToWebAudio: true
});
// Get the Web Audio Context from the plugin
var audioContext = audioinput.getAudioContext();
// Create and init a Web Audio Analyser node
analyser = audioContext.createAnalyser();
analyser.smoothingTimeConstant = 0.3;
analyser.fftSize = 1024;
processorNode = audioContext.createScriptProcessor(2048, 1, 1);
processorNode.onaudioprocess = function() {
var array = new Uint8Array(analyser.frequencyBinCount);
analyser.getByteFrequencyData(array);
var averageVolume = getAverageVolume(array)
console.log('Average volume in this chunk was: ' + averageVolume);
}
function getAverageVolume( array ) {
var amplitudeSum = 0;
for (var i = 0; i < array.length; i++) {
amplitudeSum += array[i];
}
return amplitudeSum / array.length;
}
// Connect the audioinput to the analyser node
audioinput.connect( processorNode );
If you don't want to use Web Audio, you can do the same thing using the audioinput event based method, since you'll receive chunks (arrays) of samples from the microphone, which the getAverageVolume function can handle.
I hope this helps.
Thanks for your reply.
Unfortunately, I tried your code, and it is not working.
There is no any console.log output when I give voice to the microphone.
I worked with turned streamToWebAudio to false and used the audioinput event based. However, it gave a very large time delay, so I want to try your method but not working.
Can you give me more examples?
Thanks,
Hugo Lo
Do you have some code I can look at?
Closing this now, since there haven't been any activity for a month.
I will note for future readers, this method did work for me.
Do you hear the captured sound if you connect the audioinput source or the scriptprocessor node to the audiocontext destination? I also remember something about the scriptprocessor node having to be connected to the destination or another node (that is connected to the destination) in order for it to work correctly (Chrome/Blink): http://stackoverflow.com/questions/23348109/why-must-a-scriptprocessornode-be-connected-to-a-destination
Another thing: Why does your code first fill the array with getByteTimeDomainData and then again by getByteFrequencyData?
@edimuj Thanks for your response!
Yes I hear captured sound.
I got rid of the script processor node and it works.
I was trying different graphs, now I only use frequency data thanks!
One question - How to get rid of playback of captured sound? It interferes with recording. I am not using this -
audioinput.connect(audioinput.getAudioContext().destination);
The only possible way to get this plugin to perform playback is using by the connect function you mentioned above and connecting it directly or indirectly to audioContext.destination. So if you don't use that as well as not using a scriptprocessor, I must see your code to be able to understand why audio is being played through the speakers. Anyway, since scriptprocessors need to be connected to the destination in order to work, the easiest way to silence it is to create a gain node and set it's value to zero and then connect that gain node to the destination.
Doing some housekeeping: Since there hasn't been any action regarding this issue for a time now, I'm closing it.
Thank you for your help!
I have tried to convert your pseudocode, but get some errors.
As described above, I get back only zeros in the array. From my understanding, the "processorNode" object is only responsible for repeating the volume limit. Have I understood that correctly?
Here is my code:
window.audioinput.start({
streamToWebAudio: true
});
var audioContext = window.audioinput.getAudioContext();
// Create and init a Web Audio Analyser node
var analyser = audioContext.createAnalyser();
// analyser.smoothingTimeConstant = 0.3;
analyser.fftSize = 1024;
var array = new Uint8Array(analyser.frequencyBinCount);
analyser.getByteFrequencyData(array);
console.log(array);
I get the following issue:
Uint8Array(1024) [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…]
Do you have an example for me that works, please?
In other examples, to calculate the volume level required a stream. Is there a way to get a stream through the plugin?
Example
audioContext = new AudioContext();
// ask for an audio input
navigator.getUserMedia(
{
"audio": {
"mandatory": {
"googEchoCancellation": "false",
"googAutoGainControl": "false",
"googNoiseSuppression": "false",
"googHighpassFilter": "false"
},
"optional": []
},
}, onMicrophoneGranted);
function onMicrophoneGranted(stream) {
// Create an AudioNode from the stream.
mediaStreamSource = audioContext.createMediaStreamSource(stream);
// Create a new volume meter and connect it.
meter = createAudioMeter(audioContext);
mediaStreamSource.connect(meter);
}
createAudioMeter is an added function
https://github.com/cwilso/volume-meter/blob/master/volume-meter.js
I have the problem that there is no console.log output as well.
All other function I tried are working but your example is not working:
// Start the plugin in Web Audio mode
audioinput.start({
streamToWebAudio: true
});
// Get the Web Audio Context from the plugin
var audioContext = audioinput.getAudioContext();
// Create and init a Web Audio Analyser node
analyser = audioContext.createAnalyser();
analyser.smoothingTimeConstant = 0.3;
analyser.fftSize = 1024;
processorNode = audioContext.createScriptProcessor(2048, 1, 1);
processorNode.onaudioprocess = function() {
var array = new Uint8Array(analyser.frequencyBinCount);
analyser.getByteFrequencyData(array);
var averageVolume = getAverageVolume(array)
console.log('Average volume in this chunk was: ' + averageVolume);
}
function getAverageVolume( array ) {
var amplitudeSum = 0;
for (var i = 0; i < array.length; i++) {
amplitudeSum += array[i];
}
return amplitudeSum / array.length;
}
// Connect the audioinput to the analyser node
audioinput.connect( processorNode );
Is there any solution?
Working code is:
// Global Variables for Audio
var audioContext;
var analyser;
var processorNode;
var array; // array to hold time domain data
// Start the plugin in Web Audio mode
audioinput.start({
streamToWebAudio: true
});
// Get the Web Audio Context from the plugin
audioContext = audioinput.getAudioContext();
// Create and init a Web Audio Analyser node
analyser = audioContext.createAnalyser();
analyser.smoothingTimeConstant = 0;
analyser.fftSize = 1024;
processorNode = audioContext.createScriptProcessor(2048, 1, 1);
array = new Uint8Array(analyser.frequencyBinCount);
// Connect the audioinput to the analyser node
audioinput.connect( analyser );
analyser.connect( processorNode );
processorNode.connect(audioContext.destination);
processorNode.onaudioprocess = function() {
analyser.getByteFrequencyData(array);
var averageVolume = getAverageVolume(array)
console.log('Average volume in this chunk was: ' + averageVolume);
}
function getAverageVolume( array ) {
var amplitudeSum = 0;
for (var i = 0; i < array.length; i++) {
amplitudeSum += array[i];
}
return amplitudeSum / array.length;
}
Hi desmeit, does your example process volume levels as they are happening? or only until you press stop??