How can I forward real-time transcribed audio from Azure Speech Service to an external endpoint or Azure Function?

Below is one approach using the Speech SDK’s event handlers. The SDK doesn’t offer a built‐in “pre‐routing” of recognized text to another endpoint. Instead, you can subscribe to events like Recognized and, within your event handler, asynchronously send the text (using HttpClient, for example) to your Azure Function or another endpoint before updating your UI.

For example:

CSHARPCopied
using System;
using System.Net.Http;
using System.Net.Http.Json;
using System.Threading.Tasks;
using Microsoft.CognitiveServices.Speech;

class Program
{
    static async Task Main()
    {
        // Replace with your subscription details.
        var speechConfig = SpeechConfig.FromSubscription("YourSubscriptionKey", "YourRegion");
        using var recognizer = new SpeechRecognizer(speechConfig);
        
        // Subscribe to the Recognized event.
        recognizer.Recognized += async (s, e) =>
        {
            if (e.Result.Reason == ResultReason.RecognizedSpeech)
            {
                string recognizedText = e.Result.Text;
                
                // Send the transcription to your endpoint.
                using var httpClient = new HttpClient();
                try
                {
                    await httpClient.PostAsJsonAsync("https://yourfunction.azurewebsites.net/api/YourEndpoint", 
                        new { TranscribedText = recognizedText });
                }
                catch (Exception ex)
                {
                    // Handle exceptions as needed.
                    Console.WriteLine($"Error sending text: {ex.Message}");
                }
                
                // Continue processing—for example, update your browser app's UI.
                Console.WriteLine($"Recognized: {recognizedText}");
            }
        };

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

        Console.WriteLine("Press any key to stop...");
        Console.ReadKey();

        await recognizer.StopContinuousRecognitionAsync().ConfigureAwait(false);
    }
}

Explanation

• Event Handler
  The Recognized event fires each time the service has recognized speech. Here, the event handler first checks that the result indicates recognized speech.

• Sending to an Endpoint
  Inside the handler, an asynchronous HTTP POST call sends the recognized text to your function (or any endpoint). Adjust the URL and payload to fit your scenario.

• Order of Operations
  Since you’re sending the transcription before further processing (such as updating your UI), this code gives you a chance to relay the data immediately upon recognition.

This approach provides flexibility so that you can have your browser (or backend) process the text both locally and remotely.