Edit

Share via

How to get speech to text session ID and transcription ID

  • Article

If you use speech to text and need to open a support case, you're often asked to provide a Session ID or Transcription ID of the problematic transcriptions to debug the issue. This article explains how to get these IDs.

Note

Getting Session ID

Real-time speech to text and speech translation use either the Speech SDK or the REST API for short audio.

To get the Session ID, when using SDK you need to:

  1. Enable application logging.
  2. Find the Session ID inside the log.

If you use Speech SDK for JavaScript, get the Session ID as described in this section.

If you use Speech CLI, you can also get the Session ID interactively. See details in this section.

With the speech to text REST API for short audio, you need to inject the session information in the requests. See details in this section.

Enable logging in the Speech SDK

Enable logging for your application as described in this article.

Get Session ID from the log

Open the log file your application produced and look for SessionId:. The number that would follow is the Session ID you need. In the following log excerpt example, 0b734c41faf8430380d493127bd44631 is the Session ID.

[874193]: 218ms SPX_DBG_TRACE_VERBOSE:  audio_stream_session.cpp:1238 [0000023981752A40]CSpxAudioStreamSession::FireSessionStartedEvent: Firing SessionStarted event: SessionId: 0b734c41faf8430380d493127bd44631

Get Session ID using JavaScript

If you use Speech SDK for JavaScript, you get Session ID with the help of sessionStarted event from the Recognizer class.

See an example of getting Session ID using JavaScript in this sample. Look for recognizer.sessionStarted = onSessionStarted; and then for function onSessionStarted.

Get Session ID using Speech CLI

If you use Speech CLI, then you see the Session ID in SESSION STARTED and SESSION STOPPED console messages.

You can also enable logging for your sessions and get the Session ID from the log file as described in this section. Run the appropriate Speech CLI command to get the information on using logs:

spx help recognize log

spx help translate log

Provide Session ID using REST API for short audio

Unlike Speech SDK, Speech to text REST API for short audio doesn't automatically generate a Session ID. You need to generate it yourself and provide it within the REST request.

Generate a GUID inside your code or using any standard tool. Use the GUID value without dashes or other dividers. As an example we use 9f4ffa5113a846eba289aa98b28e766f.

As a part of your REST request use X-ConnectionId=<GUID> expression. For our example, a sample request looks like this:

https://eastus.stt.speech.microsoft.com/speech/recognition/conversation/cognitiveservices/v1?language=en-US&X-ConnectionId=9f4ffa5113a846eba289aa98b28e766f

9f4ffa5113a846eba289aa98b28e766f is your Session ID.

Warning

The value of the parameter X-ConnectionId should be in the format of GUID without dashes or other dividers. All other formats aren't supported and will be discarded by the Service.

Example. If the request contains expressions like these:

  • X-ConnectionId=9f4ffa51-13a8-46eb-a289-aa98b28e766f (GUID with dividers)
  • X-ConnectionId=Request9f4ffa5113a846eba289aa98b28e766f (non-GUID)
  • X-ConnectionId=5948f700d2a811ee (non-GUID)

then the value of X-ConnectionId will not be accepted by the system, and the Session won't be found in the logs.

Getting Transcription ID for Batch transcription

Batch transcription API is a subset of the Speech to text REST API.

The required Transcription ID is the GUID value contained in the main self element of the Response body returned by requests, like Transcriptions_Create.

The following is and example response body of a Transcriptions_Create request. GUID value 537216f8-0620-4a10-ae2d-00bdb423b36f found in the first self element is the Transcription ID.

{
  "self": "https://eastus.api.cognitive.microsoft.com/speechtotext/v3.1/transcriptions/537216f8-0620-4a10-ae2d-00bdb423b36f",
  "model": {
    "self": "https://eastus.api.cognitive.microsoft.com/speechtotext/v3.1/models/base/824bd685-2d45-424d-bb65-c3fe99e32927"
  },
  "links": {
    "files": "https://eastus.api.cognitive.microsoft.com/speechtotext/v3.1/transcriptions/537216f8-0620-4a10-ae2d-00bdb423b36f/files"
  },
  "properties": {
    "diarizationEnabled": false,
    "wordLevelTimestampsEnabled": false,
    "channels": [
      0,
      1
    ],
    "punctuationMode": "DictatedAndAutomatic",
    "profanityFilterMode": "Masked"
  },
  "lastActionDateTime": "2021-11-19T14:09:51Z",
  "status": "NotStarted",
  "createdDateTime": "2021-11-19T14:09:51Z",
  "locale": "ru-RU",
  "displayName": "transcriptiontest"
}

Note

Use the same technique to determine different IDs required for debugging issues related to custom speech, like uploading a dataset using Datasets_Create request.

Note

You can also see all existing transcriptions and their Transcription IDs for a given Speech resource by using Transcriptions_Get request.