774825593db2033f45e8deecfe32c876da234415
- Introduced Qwen-Audio Realtime service for speech-to-speech processing in Pipecat. - Updated interface catalog to include Qwen-Audio Realtime capabilities. - Enhanced model resource testing to support new service. - Added configuration options for audio sample rates and turn detection modes. - Updated documentation to reflect integration details and usage instructions.
Description
No description provided
Languages
Python
47.3%
TypeScript
43%
JavaScript
8%
MDX
1.2%
CSS
0.3%
Other
0.1%