为什么即使满足了所有要求，Microsoft Cognitive Speech Api 仍发送“错误请求”

Question

我一直在尝试实现 Microsoft 认知语音转文本 API 将音频文件转换为文本，但我总是收到错误请求消息。

这是我正在实现的示例代码：

public static object MC()
{
    HttpClient client = new HttpClient();
    client.DefaultRequestHeaders.Authorization = new AuthenticationHeaderValue("Bearer", bearerToken);
    client.DefaultRequestHeaders.Add("Ocp-Apim-Subscription-Key", speechKey);
    client.DefaultRequestHeaders.Accept.Add(new MediaTypeWithQualityHeaderValue("audio/wav"));

    var mp = new MultipartFormDataContent();
    var file = new StreamContent(File.OpenRead(HttpContext.Current.Server.MapPath("~/audios/audio_06012023094443961.wav")));
    file.Headers.ContentType = new MediaTypeHeaderValue("audio/wav");
    mp.Add(file, "audio", "audio_06012023094319156.wav");

    var resp = client.PostAsync("https://" + speechRegion + ".stt.speech.microsoft.com/speech/recognition/conversation/cognitiveservices/v1?language=en-US", mp);
    while (!resp.IsCompleted)
    {
        Thread.Sleep(100);
    }
    return resp.Result.ReasonPhrase;
}

我的编程语言是C#。

Answer 1

确保包含“User-Agent”标头。由于某种未知的原因，Microsoft 认知语音 API 强制请求拥有一个。