Refactor chatbot and STT services to improve model loading, response generation, and error handling; utilize Hugging Face API for STT functionality
malek-messaoudii
e8aa76b
fix errors
malek-messaoudii
a8c8142
update stt
malek-messaoudii
47a3efb
Refactor audio processing to utilize free models and enhance logging; update TTS and STT services for improved functionality
malek-messaoudii
95cb26e
Enhance configuration validation and audio processing limits
malek-messaoudii
918acab
- Added configuration validation to ensure required environment variables are set.
- Introduced MAX_AUDIO_SIZE to limit audio file uploads to 10MB.
- Updated audio processing routes to check file size before processing.
- Adjusted TTS service to handle audio data extraction more robustly.
add files
malek-messaoudii
4a13628
Update codes
malek-messaoudii
d4b6133
Correct files
malek-messaoudii
544d113
Update tts_service.py
malek-messaoudii
4f1c42b
Update stt_service.py
malek-messaoudii
14f424f
Refactor audio models and services for improved error handling and response streaming
malek-messaoudii
9aa985d
Update gemini_client.py
malek-messaoudii
e411044
update env
malek-messaoudii
91b1985
add gemini api
malek-messaoudii
73d4f3c
Add tts/stt services
malek-messaoudii
c7fc3b6
'add'
malek-messaoudii
fd65bc8
Update training script and model files
malek-messaoudii
cfbe56e
Update training script and model files
malek-messaoudii
d9d9974
add model
malek-messaoudii
9af5350
Enhance KpaModelManager to load fine-tuned weights from Hugging Face and update requirements to include huggingface_hub
Yassine Mhirsi
d289997
Add KPA model integration and update configuration for label predictions
Yassine Mhirsi
f28285b
up
malek-messaoudii
d02227d
Add Kpa classification model
malek-messaoudii
03d23e8
Fix formatting in stance_model_manager.py by adjusting whitespace in tokenizer loading code
