azh*_*har 17 javascript node.js audiocontext web-audio-api google-cloud-speech
我正在使用带有NodeJS后端的Google Cloud API for Speech-to-text.应用程序需要能够侦听语音命令,并将它们作为缓冲区传输到后端.为此,我需要在检测到静音时发送前一个音频的缓冲区.
任何帮助,将不胜感激.包括下面的js代码
if (!navigator.getUserMedia)
navigator.getUserMedia = navigator.getUserMedia || navigator.webkitGetUserMedia ||
navigator.mozGetUserMedia || navigator.msGetUserMedia;
if (navigator.getUserMedia) {
navigator.getUserMedia({audio: true}, success, function (e) {
alert('Error capturing audio.');
});
} else alert('getUserMedia not supported in this browser.');
var recording = false;
window.startRecording = function () {
recording = true;
};
window.stopRecording = function () {
recording = false;
// window.Stream.end();
};
function success(e) {
audioContext = window.AudioContext || window.webkitAudioContext;
context = new audioContext();
// the sample rate is in context.sampleRate
audioInput = context.createMediaStreamSource(e);
var bufferSize = 4096;
recorder = context.createScriptProcessor(bufferSize, 1, 1);
recorder.onaudioprocess = function (e) {
if (!recording) return;
console.log('recording');
var left = e.inputBuffer.getChannelData(0);
console.log(convertoFloat32ToInt16(left));
};
audioInput.connect(recorder);
recorder.connect(context.destination);
}
Kai*_*ido 16
我不太确定在问题中究竟是什么问题,所以这个答案只是为了找到一种方法来检测AudioStream中的静音.
要检测AudioStream中的静音,可以使用AudioAnalyser节点,您将在该节点上getByteFrequencyData定期调用该方法,并检查在给定时间内是否有高于预期水平的声音.
您可以使用minDecibelsAnalyserNode 的属性直接设置阈值级别.
function detectSilence(
stream,
onSoundEnd = _=>{},
onSoundStart = _=>{},
silence_delay = 500,
min_decibels = -80
) {
const ctx = new AudioContext();
const analyser = ctx.createAnalyser();
const streamNode = ctx.createMediaStreamSource(stream);
streamNode.connect(analyser);
analyser.minDecibels = min_decibels;
const data = new Uint8Array(analyser.frequencyBinCount); // will hold our data
let silence_start = performance.now();
let triggered = false; // trigger only once per silence event
function loop(time) {
requestAnimationFrame(loop); // we'll loop every 60th of a second to check
analyser.getByteFrequencyData(data); // get current data
if (data.some(v => v)) { // if there is data above the given db limit
if(triggered){
triggered = false;
onSoundStart();
}
silence_start = time; // set it to now
}
if (!triggered && time - silence_start > silence_delay) {
onSoundEnd();
triggered = true;
}
}
loop();
}
function onSilence() {
console.log('silence');
}
function onSpeak() {
console.log('speaking');
}
navigator.mediaDevices.getUserMedia({
audio: true
})
.then(stream => {
detectSilence(stream, onSilence, onSpeak);
// do something else with the stream
})
.catch(console.error);而作为一个小提琴,stackSnippets可能会阻止gUM.
您可以使用SpeechRecognition result事件来确定某个单词或短语何时被识别，例如ls、cd或pwd其他命令，将.transcriptof传递SpeechRecognitionAlternative给speechSynthesis.speak()where at Attachedstart和endevent of SpeechSynthesisUtterancecall.start()或.resume()on MediaRecorderobject where MediaStreamis passed；将Blobatdataavailable事件转换为ArrayBufferusingFileReader或Response.arrayBuffer()。
我们可以选择使用audiostartor soundstartwithaudioend或soundendeventsSpeechRecognition来记录用户的实际语音，尽管末端可能与仅由标准系统麦克风捕获的音频的实际开始和结束不一致。
<!DOCTYPE html>
<html>
<head>
<title>Speech Recognition Recording</title>
</head>
<body>
<input type="button" value="Stop speech command recognition" id="stop">
<script>
navigator.mediaDevices.getUserMedia({
audio: true
})
.then(stream => {
const recorder = new MediaRecorder(stream);
const recognition = new webkitSpeechRecognition();
const synthesis = new SpeechSynthesisUtterance();
const handleResult = e => {
recognition.onresult = null;
console.log(e.results);
const result = e.results[e.results.length - 1];
if (result.isFinal) {
const [{transcript}] = result;
console.log(transcript);
synthesis.text = transcript;
window.speechSynthesis.speak(synthesis);
}
}
synthesis.onstart = () => {
if (recorder.state === "inactive") {
recorder.start()
} else {
if (recorder.state === "paused") {
recorder.resume();
}
}
}
synthesis.onend = () => {
recorder.pause();
recorder.requestData();
}
recorder.ondataavailable = async(e) => {
if (stream.active) {
try {
const blobURL = URL.createObjectURL(e.data);
const request = await fetch(blobURL);
const ab = await request.arrayBuffer();
console.log(blobURL, ab);
recognition.onresult = handleResult;
// URL.revokeObjectURL(blobURL);
} catch (err) {
throw err
}
}
}
recorder.onpause = e => {
console.log("recorder " + recorder.state);
}
recognition.continuous = true;
recognition.interimResults = false;
recognition.maxAlternatives = 1;
recognition.start();
recognition.onend = e => {
console.log("recognition ended, stream.active", stream.active);
if (stream.active) {
console.log(e);
// the service disconnects after a period of time
recognition.start();
}
}
recognition.onresult = handleResult;
stream.oninactive = () => {
console.log("stream ended");
}
document.getElementById("stop")
.onclick = () => {
console.log("stream.active:", stream.active);
if (stream && stream.active && recognition) {
recognition.abort();
recorder.stop();
for (let track of stream.getTracks()) {
track.stop();
}
console.log("stream.active:", stream.active);
}
}
})
.catch(err => {
console.error(err)
});
</script>
</body>
</html>
plnkr https://plnkr.co/edit/4DVEg6mhFRR94M5gdaIp?p=preview