Cannot Connect Locally to when using Microsoft.CognitiveServices.Speech

@Michael Lanctot Welcome to Microsoft Q&A Forum, Thank you for posting your query here!

.

• Please confirm if you are using AutoDetectSourceLanguageConfig.FromLanguages(String[]) Method to specify the possible source languages. This helps the SDK to identify the language from the provided audio.
• Did you use the RecognizeOnceAsync method to detect the language. More info here.

var result = await recognizer.RecognizeOnceAsync().ConfigureAwait(false); 
if (result.Reason == ResultReason.TranslatedSpeech) 
{ 
   var lidResult = result.Properties.GetProperty(PropertyId.SpeechServiceConnection_AutoDetectSourceLanguageResult); 
Console.WriteLine($"RECOGNIZED in '{lidResult}': Text={result.Text}");

• For some reason your code is not shared in the issue description. Please try sharing it again.

.

Awaiting your reply.

My apologies I didn't realize the code did not completely load in. Here is our code:

public async Task<string> DetectLanguageOfAudio(Stream? audioStream, CancellationToken cancellationToken)
{
    if (audioStream == null)
    {
        return string.Empty;
    }

    if (audioStream.Length == 0)
    {
        throw new InvalidOperationException("The audio stream is empty.");
    }

    var endpointUrl = new Uri(config.AzureSpeechService!.DetectLanguageEndpoint!);
    var detectConfig = SpeechConfig.FromEndpoint(endpointUrl, config.AzureSpeechService!.Key!);

    detectConfig.SetProperty(PropertyId.SpeechServiceConnection_LanguageIdMode, "Continuous");
    
    var logFilePath = @"C:\Users\mlanctot\OneDrive - United Wholesale Mortgage\Documents\Logs\speech_sdk_log.txt";
    detectConfig.SetProperty(PropertyId.Speech_LogFilename, logFilePath);

    var autoDetectSourceLanguageConfig = AutoDetectSourceLanguageConfig.FromLanguages(expectedLanguages);

    var pushStream = AudioInputStream.CreatePushStream();
    var stopRecognition = new TaskCompletionSource<string>();

    _ = Task.Run(async () =>
    {
        byte[] buffer = new byte[4096];
        int bytesRead;

        while ((bytesRead = await audioStream.ReadAsync(buffer)) > 0)
        {
            pushStream.Write(buffer, bytesRead);
        }

        pushStream.Close();
    }, cancellationToken);

    using var audioInput = AudioConfig.FromStreamInput(pushStream);
    using var recognizer = new SpeechRecognizer(detectConfig, autoDetectSourceLanguageConfig, audioInput);

    Console.WriteLine("Attempting to detect langauge");

    string detectedLanguage = string.Empty;

    recognizer.Recognizing += (s, e) =>
    {
        if (e.Result.Reason == ResultReason.RecognizingSpeech)
        {
            var autoDetectSourceLanguageResult = AutoDetectSourceLanguageResult.FromResult(e.Result);
            detectedLanguage = autoDetectSourceLanguageResult.Language; // Capture the detected language
        }
    };

    recognizer.Recognized += (s, e) =>
    {
        if (e.Result.Reason == ResultReason.RecognizedSpeech)
        {
            var autoDetectSourceLanguageResult = AutoDetectSourceLanguageResult.FromResult(e.Result);
            detectedLanguage = autoDetectSourceLanguageResult.Language; // Capture the detected language
        }
    };

    recognizer.Canceled += (s, e) =>
    {
        Console.WriteLine($"CANCELED: Reason={e.Reason}");

        if (e.Reason == CancellationReason.Error)
        {
            Console.WriteLine($"CANCELED: ErrorCode={e.ErrorCode}");
            Console.WriteLine($"CANCELED: ErrorDetails={e.ErrorDetails}");
        }

        stopRecognition.TrySetResult(detectedLanguage); // Return the detected language
    };

    recognizer.SessionStarted += (s, e) =>
    {
        Console.WriteLine("\nSession started event.");
    };

    recognizer.SessionStopped += (s, e) =>
    {
        Console.WriteLine("\nSession stopped event.");
        stopRecognition.TrySetResult(detectedLanguage); // Return the detected language
    };

    // Start continuous recognition
    await recognizer.StartContinuousRecognitionAsync().ConfigureAwait(false);

    // Wait until recognition is stopped
    await stopRecognition.Task;

    // Stop recognition
    await recognizer.StopContinuousRecognitionAsync().ConfigureAwait(false);

    audioStream.Position = 0;

    return detectedLanguage; // Return the detected language
}

Also found the log file for this as well.Cspeech_sdk_log.txt

@Michael Lanctot Thanks for sharing the details. Within the Speech SDK logs, I could see the error WS_OPEN_ERROR_UNDERLYING_IO_OPEN_FAILED which is a network related connectivity issue:

[606712]: 12842ms SPX_TRACE_ERROR: AZ_LOG_ERROR:  tlsio_openssl.c:2464 FORCE-Closing tlsio instance.

[606712]: 12844ms SPX_TRACE_SCOPE_ENTER:  uws_web_socket.cpp:247 OnWebSocketOpened

[606712]: 12845ms SPX_TRACE_ERROR:  web_socket.cpp:902 WS open operation failed with result=1(WS_OPEN_ERROR_UNDERLYING_IO_OPEN_FAILED), code=2573[0x00000a0d], time=2025-01-06T14:19:44.6496149Z

[606712]: 12845ms SPX_TRACE_INFO:  usp_connection.cpp:932 TS:270, TransportError: connection:0xa401b950, code=5, string=Connection failed (no connection to the remote host). Internal error: 1. Error details: Failed with error: WS_OPEN_ERROR_UNDERLYING_IO_OPEN_FAILED

wss://eastus2.stt.speech.microsoft.com/speech/universal/v2

X-ConnectionId: 09ce9fa5e8c14d81900fae21a6065cd1

[606712]: 12846ms SPX_TRACE_ERROR:  usp_reco_engine_adapter.cpp:1997 Response: On Error: Code:5, Message: Connection failed (no connection to the remote host). Internal error: 1. Error details: Failed with error: WS_OPEN_ERROR_UNDERLYING_IO_OPEN_FAILED

wss://eastus2.stt.speech.microsoft.com/speech/universal/v2

X-ConnectionId: 09ce9fa5e8c14d81900fae21a6065cd1.

[606712]: 12847ms SPX_DBG_TRACE_VERBOSE:  usp_reco_engine_adapter.cpp:3066 TryChangeState; audioState/uspState: 2/1000 => 2/-1 USP-ERRORERROR

[606712]: 12847ms SPX_TRACE_ERROR:  usp_reco_engine_adapter.cpp:2011 OnError: site->Error() ... error='Connection failed (no connection to the remote host). Internal error: 1. Error details: Failed with error: WS_OPEN_ERROR_UNDERLYING_IO_OPEN_FAILED

wss://eastus2.stt.speech.microsoft.com/speech/universal/v2

X-ConnectionId: 09ce9fa5e8c14d81900fae21a6065cd1'

• In Azure Portal for your Azure Speech resource, Could you please check if your Azure Speech resource --> **Networking **tab has any firewall / VNET configured ?
• Within your Corporate network, do you have any proxy / firewall ?
• Can you try using ipv4 network settings instead of ipv6 ?

Thanks for the response. I had a feeling it would be a firewall block as well but I got a response from my ticket this morning saying they don't see any block.

I also checked our resource's network tab and don't see anything that may be causing the issue.

As far as trying ipv4 instead of ipv6, how do I go about doing that?