Adding include_language_detection param to Elevenlabs Realtime STT

[mayurdd]

Adding a param to the config while connecting to the session

Please describe the changes in your PR. If it is addressing an issue, please reference that as well.

Currently, the ElevenLabsRealtimeSTTService doesn't allow us to pass include_language_detection, which can be used to get the language code in the response of the Scribe ve realtime model. After discussing with the Elevenlabs team that this is the parameter name that can be passed to get the language code, I'm creating this PR

[markbackman]

I'm happy to include this, but I can't find it in the reference docs. Can you point to where this parameter is listed in the reference docs? I'm not seeing it here:
https://elevenlabs.io/docs/api-reference/speech-to-text/v-1-speech-to-text-realtime

[mayurdd]

I'm happy to include this, but I can't find it in the reference docs. Can you point to where this parameter is listed in the reference docs? I'm not seeing it here: https://elevenlabs.io/docs/api-reference/speech-to-text/v-1-speech-to-text-realtime

Thanks for pointing out, I will flag this to their team.
The 11labs team suggested this parameter to get language_code in the response else this property will always come as None.

Below is the current response when we connect to Scribe v2 Realtime.
ElevenLabs session started: {'message_type': 'session_started', 'session_id': '15b88595fcc1dba279eefb', 'config': {'sample_rate': 8000, 'audio_format': 'pcm_8000', 'language_code': None, 'timestamps_granularity': 'word', 'vad_commit_strategy': False, 'vad_silence_threshold_secs': 1.5, 'vad_threshold': 0.4, 'min_speech_duration_ms': 100, 'min_silence_duration_ms': 100, 'max_tokens_to_recompute': 5, 'model_id': 'scribe_v2_realtime', 'disable_logging': False, 'include_timestamps': False, 'include_language_detection': False}}

Since this was launched 10 days ago, I'm sure they will be updating the docs soon to include include_language_detection and timestamps_granularity

[mayurdd]

I'm happy to include this, but I can't find it in the reference docs. Can you point to where this parameter is listed in the reference docs? I'm not seeing it here: https://elevenlabs.io/docs/api-reference/speech-to-text/v-1-speech-to-text-realtime

@markbackman Met with the 11labs team, and the documentation is updated to have that param
https://elevenlabs.io/docs/api-reference/speech-to-text/v-1-speech-to-text-realtime

[codecov]

Codecov Report

❌ Patch coverage is 0% with 3 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/pipecat/services/elevenlabs/stt.py	0.00%	3 Missing ⚠️

Files with missing lines	Coverage Δ
src/pipecat/services/elevenlabs/stt.py	0.00% <0.00%> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[markbackman]

Hey, sorry for the delay on this one.

Yes, I see this change in the docs now. Three things remain before this PR can be merged:

1. Add a docstring for the include_language_detection parameter in the the InputParams class, e.g. on L420, add:

include_language_detection: Whether to include language detection in the committed transcript.

2. Lint the code. You can run uv run scripts/fix-ruff.sh (or install the pre-commit hook: uv run pre-commit install)
3. Add a changelog fragment: https://github.com/pipecat-ai/pipecat/blob/main/CONTRIBUTING.md#changelog-entries

[markbackman]

@mayurdd if you're able, can you finish this up? If not, I'm happy to open up a branch with the changes.

[mayurdd]

@mayurdd if you're able, can you finish this up? If not, I'm happy to open up a branch with the changes.

I'll make the changes by tomorrow

[mayurdd]

@mayurdd if you're able, can you finish this up? If not, I'm happy to open up a branch with the changes.

I've updated the PR with the requested changes

[markbackman]

Thank you! 🙌