天蓝色语音到文本与音频网址

Question

您好，我已尝试使用 azure 语音转文本对音频文件进行语音转文本，一切正常。有人可以指导我如何使用音频 url 在 azure 中将语音转换为文本吗？我正在使用 REST API。

Answer 1

支持 Nicolas 所说的：如果您想使用 REST API 将音频 URL 转换为文本，批量转录是最佳选择。

获取音频的详细信息：https://learn.microsoft.com/en-US/azure/cognitive-services/speech-service/batch-transcription-audio-data

以及用于批处理的 JS 示例：https://github.com/Azure-Samples/cognitive-services-speech-sdk/tree/master/samples/batch/js/node

Answer 2

公共异步任务GenerateAudioTranscript（字符串audioFilePath） { 尝试 { var voiceConfig = SpeechConfig.FromSubscription(_configuration["SpeechCognitive:SubscriptionKey"], _configuration["SpeechCognitive:Region"]); voiceConfig.RequestWordLevelTimestamps(); voiceConfig.OutputFormat = Microsoft.CognitiveServices.Speech.OutputFormat.Detailed;

    var audioConfig = AudioConfig.FromWavFileInput(audioFilePath);
    var recognizer = new SpeechRecognizer(speechConfig, audioConfig);

    var speechRecognizerWaiter = new TaskCompletionSource<string>();
    var transcript = new List<AudioTranscriptDto>();
    int sequenceNumber = 1;
    recognizer.Recognized += (s, e) =>
    {
        if (e.Result.Reason == ResultReason.RecognizedSpeech)
        {
            string text = e.Result.Text;
            var duration = e.Result.Duration;
            long offset = e.Result.OffsetInTicks;

            var startTime = TimeSpan.FromTicks(offset);
            var endTime = startTime + duration;
            var srtBlock = $"{sequenceNumber}\n{FormatTimeSpan(startTime)} --> {FormatTimeSpan(endTime)}\n{text}\n\n";

            transcript.Add(new AudioTranscriptDto
            {
                SequenceNumber = sequenceNumber,
                StartTime = FormatTimeSpan(startTime),
                EndTime = FormatTimeSpan(endTime),
                Text = text
            });
            sequenceNumber++;
        }
    };

    recognizer.SessionStarted += (sender, e) =>
    {
        Console.WriteLine("-----------> started");
    };

    recognizer.SessionStopped += (sender, e) =>
    {
        Console.WriteLine("-----------> stooped");
        speechRecognizerWaiter.SetResult("Recognition completed");
    };

    await recognizer.StartContinuousRecognitionAsync();
    var str = await speechRecognizerWaiter.Task;
    await recognizer.StopContinuousRecognitionAsync();

    return transcript;
}
catch (Exception ex)
{

    throw;
}

}

静态字符串FormatTimeSpan(TimeSpan timeSpan) { return $"{timeSpan.Hours:D2}:{timeSpan.Minutes:D2}:{timeSpan.Seconds:D2},{timeSpan.Milliseconds:D3}"; }在此输入代码