so recently for a work project I've been playing around with speech to text models and in particular custom speech to text models. With a bit of mixing and matching examples I've managed to get a test application to talk to the normal Bing speech to text API. But when I attempt to use it with a custom speech instance only the HTTPS URL works. When I use any of the available long form web socket URLS the error An unhandled exception of type 'System.NullReferenceException' occurred in SpeechClient.dll occurs. This is a bit of a problem as that endpoint only supports 2 minutes of transcription, where as the websocket endpoint supports up to 10 minutes.
This https://learn.microsoft.com/en-us/azure/cognitive-services/custom-speech-service/customspeech-how-to-topics/cognitive-services-custom-speech-use-endpoint page here is what I'm going off of. It says that I should use a web socket url when creating the service, but that leads to the error above.
Here my test bed code for trying it out:
using System;
using Microsoft.CognitiveServices.SpeechRecognition;
using System.IO;
namespace ConsoleApp1
{
class Program
{
DataRecognitionClient dataClient;
static void Main(string[] args)
{
Program p = new Program();
p.Run(args);
}
void Run(string[] args)
{
try
{
// Works
//this.dataClient = SpeechRecognitionServiceFactory.CreateDataClient(SpeechRecognitionMode.LongDictation, "en-US", "Key");
// Works
//this.dataClient = SpeechRecognitionServiceFactory.CreateDataClient(SpeechRecognitionMode.LongDictation, "en-US",
// "Key", "Key",
// "https://Id.api.cris.ai/ws/cris/speech/recognize/continuous");
// Doesn't work
this.dataClient = SpeechRecognitionServiceFactory.CreateDataClient(SpeechRecognitionMode.LongDictation, "en-US",
"Key", "Key",
"wss://Id.api.cris.ai/ws/cris/speech/recognize/continuous");
this.dataClient.AuthenticationUri = "https://westus.api.cognitive.microsoft.com/sts/v1.0/issueToken";
this.dataClient.OnResponseReceived += this.ResponseHandler;
this.dataClient.OnConversationError += this.ErrorHandler;
this.dataClient.OnPartialResponseReceived += this.PartialHandler;
Console.WriteLine("Starting Transcription");
this.SendAudioHelper("Audio file path");
(new System.Threading.ManualResetEvent(false)).WaitOne();
} catch(Exception e)
{
Console.WriteLine(e);
}
}
private void SendAudioHelper(string wavFileName)
{
using (FileStream fileStream = new FileStream(wavFileName, FileMode.Open, FileAccess.Read))
{
// Note for wave files, we can just send data from the file right to the server.
// In the case you are not an audio file in wave format, and instead you have just
// raw data (for example audio coming over bluetooth), then before sending up any
// audio data, you must first send up an SpeechAudioFormat descriptor to describe
// the layout and format of your raw audio data via DataRecognitionClient's sendAudioFormat() method.
int bytesRead = 0;
byte[] buffer = new byte[1024];
try
{
do
{
// Get more Audio data to send into byte buffer.
bytesRead = fileStream.Read(buffer, 0, buffer.Length);
// Send of audio data to service.
this.dataClient.SendAudio(buffer, bytesRead);
}
while (bytesRead > 0);
}
finally
{
// We are done sending audio. Final recognition results will arrive in OnResponseReceived event call.
this.dataClient.EndAudio();
}
}
}
void ErrorHandler(object sender, SpeechErrorEventArgs e)
{
Console.WriteLine(e.SpeechErrorText);
}
void ResponseHandler(object sender, SpeechResponseEventArgs e)
{
if(e.PhraseResponse.RecognitionStatus == RecognitionStatus.EndOfDictation || e.PhraseResponse.RecognitionStatus == RecognitionStatus.DictationEndSilenceTimeout)
{
Console.WriteLine("Trnascription Over");
Console.ReadKey();
Environment.Exit(0);
}
for(int i = 0; i < e.PhraseResponse.Results.Length; i++)
{
Console.Write(e.PhraseResponse.Results[i].DisplayText);
}
Console.WriteLine();
}
void PartialHandler(object sender, PartialSpeechResponseEventArgs e)
{
}
}
}
Thanks in advance for any help.
