Merge pull request #2612 from pipecat-ai/filipi/deepgram_flux

Integrating the new Deepgram model (Flux) with Pipecat
This commit is contained in:
Filipi da Silva Fuchter
2025-10-01 21:20:47 -03:00
committed by GitHub
5 changed files with 797 additions and 0 deletions

View File

@@ -9,6 +9,14 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
### Added
- Added `WebsocketSTTService` base class for websocket-based STT services.
Combines STT functionality with websocket connectivity, providing automatic
error handling and reconnection capabilities with exponential backoff.
- Added `DeepgramFluxSTTService` for real-time speech recognition using
Deepgram's Flux WebSocket API. Flux understands conversational flow and
automatically handles turn-taking.
- Added RTVI messages for user/bot audio levels and system logs.
- Include OpenAI-based LLM services cached tokens to `MetricsFrame`.