Live Streaming Audio Transcription

The CreateLiveTranscriptionClient method encapsulates a websocket connection to the Deepgram API and returns a class that emits events received from the Deepgram API for consumption by your application.

ℹ️

LiveTranscriptionClient Lifecycle

The LiveTranscriptionClient is Disposable so it should always be created in a 'using' to ensure it is disposed of properly.

Initiating a Connection

The StartConnectionAsync method initiates a connection with the Deepgram API via websocket. You may pass in options for the transcription as well.

using (var deepgramLive = deepgram.CreateLiveTranscriptionClient())
{
  var options = new LiveTranscriptionOptions()
  {
    Punctuate = true,
    Diarize = true,
    Encoding = Deepgram.Common.AudioEncoding.Linear16
  };
  await deepgramLive.StartConnectionAsync(options);
}

Events

The following events are fired by the LiveTranscriptionClient:

EventDescriptionData
ConnectionOpenedThe websocket connection to Deepgram has been opened.The DG live transcription object
ConnectionClosedThe websocket connection to Deepgram has been closed.WebSocket.CloseEvent
ConnectionErrorAn error occurred with the websocket connectionError object
TranscriptReceivedDeepgram has responded with a transcriptionTranscription response

ConnectionOpened

Fires when the WebSocket connection to Deepgram has been opened. Ensure you begin sending your audio data once the connection to Deepgram has been established.

ℹ️

Note

The example below demonstrates sending a pre-recorded audio to simulate a real-time stream of audio. In a real application, this type of audio is better handled using the pre-recorded transcription.

using (var deepgramLive = deepgram.CreateLiveTranscriptionClient())
{
    deepgramLive.ConnectionOpened += HandleConnectionOpened;

    // Connection opened so start sending audio.
    void HandleConnectionOpened(object? sender, ConnectionOpenEventArgs e)
    {
        byte[] buffer;

        using (FileStream fs = File.OpenRead("C:\\Users\\Michael\\Desktop\\Bueller.wav"))
        {
            buffer = new byte[fs.Length];
            fs.Read(buffer, 0, (int)fs.Length);
        }

        var chunks = buffer.Chunk(1000);

        foreach (var chunk in chunks)
        {
            deepgramLive.SendData(chunk);
            Task.Delay(50).Wait();
        }

        deepgramLive.FinishAsync();
    }
}

ConnectionClosed

Fires when the WebSocket connection is closed.

using (var deepgramLive = deepgram.CreateLiveTranscriptionClient())
{
    deepgramLive.ConnectionClosed += HandleConnectionClosed;

    void HandleConnectionClosed(object? sender, ConnectionClosedEventArgs e)
    {
        Console.Write("Connection Closed");
    }
}

ConnectionError

Fires on any error in the connection, sending or receiving.

using (var deepgramLive = deepgram.CreateLiveTranscriptionClient())
{
    deepgramLive.ConnectionError += HandleConnectionError;

    void HandleConnectionError(object? sender, ConnectionErrorEventArgs e)
    {
        Console.WriteLine(e.Exception.Message);
    }
}

TranscriptReceived

Fires when a transcript is received from the Deepgram API.

using (var deepgramLive = deepgram.CreateLiveTranscriptionClient())
{
    deepgramLive.TranscriptReceived += HandleTranscriptReceived;

    void HandleTranscriptReceived(object? sender, TranscriptReceivedEventArgs e)
    {
        if (e.Transcript.IsFinal &&
            e.Transcript.Channel.Alternatives.First().Transcript.Length > 0) {
            var transcript = e.Transcript;
            Console.WriteLine($"{transcript.Channel.Alternatives.First().Transcript}");
        }
    }
}

Functions

The object returned by the transcription.live function provides several functions to make using the Deepgram API easier. They are send, getReadyState, and finish.

Sending Data

The SendData method sends raw audio data to the Deepgram API.

deepgramLive.SendData(AUDIO_STREAM_DATA);

Get Websocket State

The State method returns the ready state of the websocket connection to Deepgram.

var websocketReadyState = deepgramLive.State();

Finish Transcribing

The FinishAsync method sends a signal to the Deepgram API that you have finished streaming audio and it should conclude its transcription efforts. Upon sending the final transcript back it will close the websocket connection.

deepgramLive.FinishAsync()

Closing the Connection

The StopConnectionAsync method closes the Websocket connection to Deepgram.

⚠️

You May Want FinishAsync Instead

In most instances, the FinishAsync method should be used to end a streaming session. Otherwise, the final transcript will not be received. Using the FinishAsync method will also cause the connection to close.

deepgramLive.StopConnectionAsync()