Microsoft Speech-to-Text SDK JS 不接受包含长字节数组的文件

Question

我正在使用 Microsoft 的 Azure 语音到文本 SDK 从 a.wav 文件中获取文本，使用 JavaScript。 问题是，识别器不会接受文件 object 并返回错误“Uncaught rangeerror: source array is too long”。 在用于使文件 object 正常工作的 blob 上调用 .slice(0, 2248)，返回 .wav 文件的正确第一个字。 但是，如果我尝试将 blob 切成块，例如 (2249, 4497)，则会返回错误“Uncaught rangeerror: offset is outside the bounds of the DataView”。 我不知道如何 a) 让识别器接受具有长源数组的 blob 或 b) 将 blob 分成不超出范围的块。 .wav url 已更改为匿名破折号，应忽略。 任何解决方案表示赞赏！

JS:
<script>
//get wav file from url, create File object with it
function fromFile() {
  fetch("http://www.-----------.com/prod/wp-content/uploads/2020/12/cutafew.wav")
  .then(response => response.blob())
  .then(blob => {
    var file = new File([blob], "http://www.---------.com/prod/wp-content/uploads/2020/12/cutafew.wav", {
               type:"audio/x-wav", lastModified:new Date().getTime()
             });

//if file got successfully, do the following:
    var reader = new FileReader();
    var speechConfig = SpeechSDK.SpeechConfig.fromSubscription("f6abc3bfabc64f0d820d537c0d738788", "centralus");
    var audioConfig = SpeechSDK.AudioConfig.fromWavFileInput(file);
    var recognizer = new SpeechSDK.SpeechRecognizer(speechConfig, audioConfig);


//use recognizer to convert wav file to text
recognizer.recognizing = (s, e) => {
console.log(e.result);
    console.log(`RECOGNIZING: Text=${e.result.text}`);
};

recognizer.recognized = (s, e) => {
    if (e.result.reason == ResultReason.RecognizedSpeech) {
        console.log(`RECOGNIZED: Text=${e.result.text}`);
    }
    else if (e.result.reason == ResultReason.NoMatch) {
        console.log("NOMATCH: Speech could not be recognized.");
    }
};

recognizer.canceled = (s, e) => {
    console.log(`CANCELED: Reason=${e.reason}`);

    if (e.reason == CancellationReason.Error) {
        console.log(`"CANCELED: ErrorCode=${e.errorCode}`);
        console.log(`"CANCELED: ErrorDetails=${e.errorDetails}`);
        console.log("CANCELED: Did you update the subscription info?");
    }

    recognizer.stopContinuousRecognitionAsync();
};

recognizer.sessionStopped = (s, e) => {
    console.log("\n    Session stopped event.");
    recognizer.stopContinuousRecognitionAsync();
};

recognizer.startContinuousRecognitionAsync();

})
//throw error if file wasn't created
.catch(err => console.error(err));

}

fromFile();
</script>

Answer 1

您可以使用从内存 stream 示例中识别

const fs = require('fs');
const sdk = require("microsoft-cognitiveservices-speech-sdk");
const speechConfig = sdk.SpeechConfig.fromSubscription("<paste-your-speech-key-here>", "<paste-your-speech-location/region-here>");

function fromStream() {
    let pushStream = sdk.AudioInputStream.createPushStream();

    fs.createReadStream("YourAudioFile.wav").on('data', function(arrayBuffer) {
        pushStream.write(arrayBuffer.slice());
    }).on('end', function() {
        pushStream.close();
    });
 
    let audioConfig = sdk.AudioConfig.fromStreamInput(pushStream);
    let recognizer = new sdk.SpeechRecognizer(speechConfig, audioConfig);
    recognizer.recognizeOnceAsync(result => {
        console.log(`RECOGNIZED: Text=${result.text}`);
        recognizer.close();
    });
}
fromStream();