JS 录制压缩16K采样率(SampleRate)音频

不会跳舞的鸟

发布于 2022-11-16 09:41:31

2.5K0

发布于 2022-11-16 09:41:31

文章被收录于专栏：不会跳舞的鸟

背景

考研复习期间压力大，写点代码放松一下。另外，没接触过音频编码相关知识…

开搞

目的

涉及到 JS 的语音识别，录音我使用的是 Recorder.js。各个平台只接受 16K/8K SampleRate 的音频，JS 录制默认是 44.1K SampleRate，这时需要压缩采样率。

可选方案

ffmpeg.js（ffmpeg 的 JS 移植版），大略看一下，用起来貌似有点麻烦，库还很大，暂且放弃。
手动压。

手动压

参考了 http://www.cnblogs.com/blqw/p/3782420.html ，可是目的不完全相同，于是我在它的基础上操刀。主要针对两个函数入手：interleave、encodeWAV。首先在上部分声明一个“newSampleRate”

var newSampleRate = 16000;

原版

// 原版
function interleave(inputL, inputR) {
    var length = inputL.length + inputR.length;
    var result = new Float32Array(length);

    var index = 0,
        inputIndex = 0;

    while (index < length) {
        result[index++] = inputL[inputIndex];
        result[index++] = inputR[inputIndex];
        inputIndex++;
    }
    return result;
}

function encodeWAV(samples) {
    var buffer = new ArrayBuffer(44 + samples.length * 2);
    var view = new DataView(buffer);

    /* RIFF identifier */
    writeString(view, 0, 'RIFF');
    /* RIFF chunk length */
    view.setUint32(4, 36 + samples.length * 2, true);
    /* RIFF type */
    writeString(view, 8, 'WAVE');
    /* format chunk identifier */
    writeString(view, 12, 'fmt ');
    /* format chunk length */
    view.setUint32(16, 16, true);
    /* sample format (raw) */
    view.setUint16(20, 1, true);
    /* channel count */
    view.setUint16(22, numChannels, true);
    /* sample rate */
    view.setUint32(24, sampleRate, true);
    /* byte rate (sample rate * block align) */
    view.setUint32(28, sampleRate * 4, true);
    /* block align (channel count * bytes per sample) */
    view.setUint16(32, numChannels * 2, true);
    /* bits per sample */
    view.setUint16(34, 16, true);
    /* data chunk identifier */
    writeString(view, 36, 'data');
    /* data chunk length */
    view.setUint32(40, samples.length * 2, true);

    floatTo16BitPCM(view, 44, samples);

    return view;
}
}, self);

修改后单声道版（大部分语音识别都只支持单声道）

// 修改后 单声道
function interleave(inputL) {
    var compression = 44100 / newSampleRate;
    var length = inputL.length / compression;
    console.log('length:' + length)
    var result = new Float32Array(length);

    var index = 0,
        inputIndex = 0; 

    while (index < length){
        result[index++] = inputL[parseInt(inputIndex)];
        inputIndex += compression;
    }
    return result;
}

// 修改后 单声道版
function encodeWAV(samples) {
    var buffer = new ArrayBuffer(44 + samples.length * 2);
    var view = new DataView(buffer);

    numChannels = 1;

    let bitPerSample = 16 * 2,
        blockAlign = numChannels * (bitPerSample / 8);

    /* RIFF identifier */
    writeString(view, 0, 'RIFF');
    /* RIFF chunk length */
    view.setUint32(4, 36 + samples.length * 2, true);
    /* RIFF type */
    writeString(view, 8, 'WAVE');
    /* format chunk identifier */
    writeString(view, 12, 'fmt ');
    /* format chunk length */
    view.setUint32(16, 16, true);
    /* sample format (raw) */
    view.setUint16(20, 1, true);
    /* channel count */
    view.setUint16(22, numChannels, true);
    /* sample rate */
    view.setUint32(24, newSampleRate, true);
    /* byte rate (sample rate * block align) */
    view.setUint32(28, newSampleRate * blockAlign, true);
    /* block align (channel count * bytes per sample) */
    view.setUint16(32, blockAlign, true);
    /* bits per sample */
    view.setUint16(34, bitPerSample, true);
    /* data chunk identifier */
    writeString(view, 36, 'data');
    /* data chunk length */
    view.setUint32(40, samples.length * 2, true);

    floatTo16BitPCM(view, 44, samples);

    return view;
}

修改后双声道版

// 修改后 双声道版
function interleave(inputL, inputR) {
    var compression = 44100 / newSampleRate; // 
    var length = (inputL.length + inputR.length) / compression;
    console.log('length:' + length)
    var result = new Float32Array(length);

    var index = 0,
        inputIndex = 0; 

    while (index < length){
        result[index++] = inputL[parseInt(inputIndex)];
        result[index++] = inputR[parseInt(inputIndex)];
        inputIndex += compression;
    }
    return result;
}

// 修改后 双声道版
function encodeWAV(samples) {
    var buffer = new ArrayBuffer(44 + samples.length * 2);
    var view = new DataView(buffer);

    numChannels = 2;

    let bitPerSample = 16,
        blockAlign = numChannels * (bitPerSample / 8);

    /* RIFF identifier */
    writeString(view, 0, 'RIFF');
    /* RIFF chunk length */
    view.setUint32(4, 36 + samples.length * 2, true);
    /* RIFF type */
    writeString(view, 8, 'WAVE');
    /* format chunk identifier */
    writeString(view, 12, 'fmt ');
    /* format chunk length */
    view.setUint32(16, 16, true);
    /* sample format (raw) */
    view.setUint16(20, 1, true);
    /* channel count */
    view.setUint16(22, numChannels, true);
    /* sample rate */
    view.setUint32(24, newSampleRate, true);
    /* byte rate (sample rate * block align) */
    view.setUint32(28, newSampleRate * blockAlign, true);
    /* block align (channel count * bytes per sample) */
    view.setUint16(32, blockAlign, true);
    /* bits per sample */
    view.setUint16(34, bitPerSample, true);
    /* data chunk identifier */
    writeString(view, 36, 'data');
    /* data chunk length */
    view.setUint32(40, samples.length * 2, true);

    floatTo16BitPCM(view, 44, samples);

    return view;
}
}, self);

不过最后效果好像还有些缺陷…不保证能用…以后有时间再研究研究…

本文参与腾讯云自媒体同步曝光计划，分享自作者个人站点/博客。

如有侵权请联系 cloudcommunity@tencent.com 删除

视频处理