# Update service (/en/realtime-media/speech-to-text/build/start-transcribing-and-translating/update-service)

> For AI agents: see the complete documentation index at [llms.txt](/llms.txt).

Real-Time STT supports the `update` command, which allows modification of parameters without stopping and restarting the service. For example, when transcription is enabled for a specified host only, and this host disconnects and rejoins the channel with a different UID, the Real-Time STT service needs to subscribe to the new UID and transcribe their audio. The `update` command can help resolve this problem.

The `update` command scope includes the following:

* Update transcription languages.
* Update UIDs of specified hosts whose audio needs to be transcribed.
* Switch between transcribing all hosts and a specified host.

## Prerequisites [#prerequisites]

To follow this procedure, you must:

* Have a valid [Agora Account](https://console.agora.io/v2).

* Have a valid Agora project with an app ID and a temporary token or a token
  server. For details, see [Agora account management](/en/realtime-media/voice/manage-agora-account).

* Have a computer with access to the internet. If your network has a firewall, follow the steps in [Firewall requirements](../../reference/firewall).

* Join a Video SDK channel as a host and start streaming. Refer to the [Voice SDK quickstart](/en/realtime-media/voice/quickstart) guide.

* Make sure Real-Time STT is enabled for your app.

## Implementation [#implementation]

Follow the API call sequence from the [REST Quickstart](../../get-started/quickstart).

### Request examples [#request-examples]

* Update the transcription language:

  ```bash
  curl --location --request POST 'https://api.agora.io/api/speech-to-text/v1/projects/{appId}/agents/{agentId}/update?sequenceId=1&updateMask=languages' \
  --header 'Content-Type: application/json' \
  --header 'Authorization: Basic <credentials>' \
  --data '{
      "languages": [
          "en-US",
          "ru-RU"
      ]
  }'
  ```

* Update the UID

  ```bash
  curl --location --request POST 'https://api.agora.io/api/speech-to-text/v1/projects/{appId}/agents/{agentId}/update?sequenceId=1&updateMask=rtcConfig.subscribeAudioUids' \
  --header 'Content-Type: application/json' \
  --header 'Authorization: Basic <credentials>' \
  --data '{
      "rtcConfig": {
          "subscribeAudioUids": ["12", "23", "45"],
      }
  }'
  ```

### Response examples [#response-examples]

* Success

  ```json
  {
    "agent_id": "Agent ID.",
    "create_ts": 12345678,
    "status": "RUNNING"
  }
  ```

* Failure

  ```json
  {
   "message": "string" // error reason
  }
  ```

### Considerations [#considerations]

* Calling `update` requires at least 5-second intervals, specifically:
  * After calling `start`, the app needs to wait at least 5 seconds to call the `update` command.
  * After calling `update`, the app needs to wait at least 5 seconds before calling `update` again.
  * There is no interval limitation for calling `update` with `query` and `stop` commands.
* When updated, the new parameter fully replaces the old parameter, unless the old parameter is masked with `updateMask`. For example, say the audio of hosts with UID 100 and 200 is being transcribed. The host with the UID 200 disconnects and rejoins with the new UID 300. The `update` command needs to use `"subscribeAudioUids": "[ "100", "300" ]`. If it uses only `"subscribeAudioUids": "[ "300" ]`, then only the UID 300 host's audio will be transcribed. The languages configuration will not be updated.
* If languages and host UIDs are updated together, pass in `"updateMask=languages,rtcConfig.subscribeAudioUids"`.
* `sequenceId` should increase by 1 for each request.