Question

我想从ArrayBuffer播放音频数据......所以我生成了我的数组并用微缩输入填充它。如果我在画布上绘制这些数据，它看起来像 - ＆gt; enter image description here

这样可行！

但如果我想用

收听这些数据

context.decodeAudioData(tmp, function(bufferN) { //tmp is a arrayBuffer
    var out = context.createBufferSource();
    out.buffer = bufferN;
    out.connect(context.destination);
    out.noteOn(0);
}, errorFunction);

我听不到任何声音......因为调用了errorFunction。但是错误是空的！

我也尝试过这样的缓冲区：

var soundBuffer = context.createBuffer(myArrayBuffer, true/*make mono*/);

但我收到错误：Uncaught SyntaxError：指定了无效或非法字符串。

是谁能给我一个提示？

编辑1（更多代码以及我如何获得麦克风输入）：

 navigator.webkitGetUserMedia({audio: true}, function(stream) {

                liveSource = context.createMediaStreamSource(stream);

                // create a ScriptProcessorNode
                if(!context.createScriptProcessor){
                   node = context.createJavaScriptNode(2048, 1, 1);
                } else {
                   node = context.createScriptProcessor(2048, 1, 1);
                }


                node.onaudioprocess = function(e){

               var tmp = new Uint8Array(e.inputBuffer.byteLength);
               tmp.set(new      Uint8Array(e.inputBuffer.byteLength), 0);

   //Here comes the code from above.

感谢您的帮助！

Answer 1

回调函数返回的错误为null，因为在当前webaudio api spec中该函数没有返回对象错误

callback DecodeSuccessCallback = void (AudioBuffer decodedData);
callback DecodeErrorCallback = void ();

    void decodeAudioData(ArrayBuffer audioData,
                         DecodeSuccessCallback successCallback,
                         optional DecodeErrorCallback errorCallback);

当完整的输入ArrayBuffer被解码并在内部存储为AudioBuffer时会引发DecodeSuccessCallback但由于某些未知原因，decodeAudioData无法解码实时流。

处理音频时，您可以尝试播放捕获的缓冲区设置输出缓冲区数据

function connectAudioInToSpeakers(){

  //var context = new webkitAudioContext();  
  navigator.webkitGetUserMedia({audio: true}, function(stream) {

    var context = new webkitAudioContext();  
    liveSource = context.createMediaStreamSource(stream);

    // create a ScriptProcessorNode
    if(!context.createScriptProcessor){
       node = context.createJavaScriptNode(2048, 1, 1);
    } else {
       node = context.createScriptProcessor(2048, 1, 1);
    }


    node.onaudioprocess = function(e){

        try{
            ctx.clearRect(0, 0, document.getElementById("myCanvas").width, document.getElementById("myCanvas").height);
            document.getElementById("myCanvas").width = document.getElementById("myCanvas").width;
            ctx.fillStyle="#FF0000";

            var input = e.inputBuffer.getChannelData(0);
            var output = e.outputBuffer.getChannelData(0);
            for(var i in input) {
                output[i] = input[i];
                ctx.fillRect(i/4,input[i]*500+200,1,1);
            }


        }catch (e){
            console.log('node.onaudioprocess',e.message);
        }

    }

     // connect the ScriptProcessorNode with the input audio
    liveSource.connect(node);
    // if the ScriptProcessorNode is not connected to an output the "onaudioprocess" event is not triggered in chrome
    node.connect(context.destination);

    //Geb mic eingang auf boxen
    //liveSource.connect(context.destination);
  });
}

Answer 2

过了一会儿，我试图再次解决这个问题并找到了解决方案：

https://developer.mozilla.org/en-US/docs/Web/API/ScriptProcessorNode

它根本不复杂，所以我创造了一个工作小提琴：

https://jsfiddle.net/WEM3y/

所以激活你的麦克风（在chrome v35上测试）并检查出来。

我改变了部分：

node.onaudioprocess = function(e){

    var outData = e.outputBuffer.getChannelData(0);
    var inData = e.inputBuffer.getChannelData(0);

    // Loop through the 4096 samples, copy them to output buffer
    for (var sample = 0; sample < e.outputBuffer.length; sample++) {
      // Set the data in the output buffer for each sample
      outData[sample] = inData[sample]; //Modify your buffer here if you want
    }
}

Answer 3

您的上下文初始化可能是按照OJay在Why does this code work in Safari but not Chrome? Arrrgh

中的建议完成的

decodeAudioData HTML5音频API

3 个答案: