CODEWITHSUNDEEP
google-cloud-platformspeech-recognitionspeech-to-text
Simon
Simon

Reputation: 207

Got error message "The caller does not have permission" when using RecognitionAudio.FromStream()

I have successfully transcribed a audio file with the google Speech-to-Text API from my code when I used fetchFromUri like this code:

var file = RecognitionAudio.fetchFromUri(fileUri);
var longRunningRecognizeOperation = speech.LongRunningRecognize(config, file);

and also when is use fetchFromStorageUri like this:

var storageFile = RecognitionAudio.fetchFromStorageUri(gsStorageUri);
var longRunningRecognizeOperation = speech.LongRunningRecognize(config, storageFile);

But as soon I try to use it with fromStream

var audioStream = RecognitionAudio.fromStream(fileStream);
var longRunningRecognizeOperation = speech.LongRunningRecognize(config, audioStream);

I got the error

"The caller does not have permission"

We use a service account key json in our project but it seems that we have t o configure some permission to get RecognitionAudio.fromStream work.

Does anyone know what to do? :)

Upvotes: 1

Views: 229

Answers (1)

Simon
Simon

Reputation: 207

The error message was misleading as it did indicate a permission issue. Instead, the error was related to the stream. However, I was able to resolve the issue by converting the stream into a ByteArray.

Like this:

using (MemoryStream memoryStream = new MemoryStream())
{
    stream.Seek(0, SeekOrigin.Begin);
    stream.CopyTo(memoryStream);
    var audioByteArray = memoryStream.ToArray();
    RecognitionAudio recognitionAudio = RecognitionAudio.FromBytes(audioByteArray);

    var longRunningRecognizeOperation = speech.LongRunningRecognize(googleTranscriptionConfig, recognitionAudio);
    longRunningRecognizeOperation = longRunningRecognizeOperation.PollUntilCompleted();

    var response = longRunningRecognizeOperation.Result;
    return response;
}

Upvotes: 0

Related Questions