public string ConvertSpeechToText(byte[] bytes)
{
System.Environment.SetEnvironmentVariable("GOOGLE_APPLICATION_CREDENTIALS", "Speech-206301277cf5.json");
var speech = SpeechClient.Create();
var response = speech.Recognize(new RecognitionConfig()
{
Encoding = RecognitionConfig.Types.AudioEncoding.Linear16,
LanguageCode = "en",
}, RecognitionAudio.FromBytes(bytes));
string text = "";
foreach (var result in response.Results)
{
foreach (var alternative in result.Alternatives)
{
text += alternative.Transcript;
}
}
return(text);
}
object AsyncRecognizeGcs(string storageUri)
{
try
{
var speech = SpeechClient.Create();
var longOperation = speech.LongRunningRecognize(new RecognitionConfig()
{
Encoding = RecognitionConfig.Types.AudioEncoding.Linear16,
Model = voiceModelDropdown.SelectedItem.ToString(),
EnableAutomaticPunctuation = enableAutoPunctuationDropdown.SelectedItem.ToString() == "Yes" ? true : false, //true if yes selected
// SampleRateHertz = processor.getSampleRate(filePath), //another customisable option
LanguageCode = languageCodeBox.Text.ToString(),
}, RecognitionAudio.FromStorageUri(storageUri));
longOperation = longOperation.PollUntilCompleted();
var response = longOperation.Result;
foreach (var result in response.Results)
{
foreach (var alternative in result.Alternatives)
{
richTextBox1.Text = alternative.Transcript;
richTextBox1.BeginInvoke((MethodInvoker) delegate { richTextBox1.AppendText(alternative.Transcript + "\n"); }); //delegate to write to a control from different thread
transcribedText += alternative.Transcript + "\n";
Console.WriteLine($"Transcript: { alternative.Transcript}");
}
}
processingLabel.Visible = false;
saveTranscriptBtn.Enabled = true;
}
catch (Exception ex)
{
MessageBox.Show("An error occured, see the cloud error reporting page for details");
cloudErrors.report(ex);
}
return(0);
}
private void transcribeBtn_Click(object sender, EventArgs e)
{
processingLabel.Visible = true;
//need to add some sort of loading screen cause this takes a bit of time
if (filePath.Contains(".mp3"))
{
//convert mp3 to wav - not required currently
}
newFilePath = filePath + "m.wav"; //have to create new file, but it will be deleted after it's used
if (URIString == "")
{
if (audioProcessor.getDuration(filePath) <= new TimeSpan(0, 1, 0))
{
audioProcessor.ConvertStereoToMono(filePath, newFilePath); //usually files are recorded stereo, we need mono here
var speech = SpeechClient.Create();
var response = speech.Recognize(new RecognitionConfig()
{
Encoding = RecognitionConfig.Types.AudioEncoding.Linear16,
Model = voiceModelDropdown.SelectedItem.ToString(),
EnableAutomaticPunctuation = enableAutoPunctuationDropdown.SelectedItem.ToString() == "Yes" ? true : false, //true if yes selected
// SampleRateHertz = processor.getSampleRate(filePath), //another customisable option
LanguageCode = languageCodeBox.Text.ToString(),
//maxAlternatives, profanityFilter, speechContext
}, RecognitionAudio.FromFile(newFilePath));
foreach (var result in response.Results)
{
foreach (var alternative in result.Alternatives)
{
richTextBox1.Text = alternative.Transcript;
transcribedText += alternative.Transcript;
}
}
saveTranscriptBtn.Enabled = true;
processingLabel.Visible = false;
}
else //if file is longer than 1 minute
{
MessageBox.Show("File is longer than 1 minute, please upload the file to Google Cloud and use an URI to access it");
transcribeBtn.Enabled = false;
processingLabel.Visible = false;
}
}
else //if URI string has been provided
{
try
{
// asyncTranscribe(URIString);
AsyncRecognizeGcs(URIString);
//"gs://s2t-test-bucket1/speech.mp3m.wav"
}
catch (UriFormatException UriException)
{
//write to an error log maybe
MessageBox.Show("Invalid URI string");
cloudErrors.report(UriException);
}
}
File.Delete(newFilePath);
}
