Fix TTS language issue by dynamically updating model in /translate-audio endpoint

Removed webrtcvad to avoid the build error.

Replaced librosa with torchaudio for audio loading and resampling. Added speech detection (energy-based or webrtcvad for accuracy). Improved /translate-audio endpoint to handle silent audio gracefully.

Updated title to Talklas API. Added description of /translate-audio endpoint with speech detection feature. Included Features section (speech detection, supported models) and Supported Languages section.

Added apt-get install for libsndfile1 (required by soundfile) and ffmpeg (required by torchaudio).

Replaced librosa with torchaudio; removed librosa and numba. Added torchaudio (audio processing) and webrtcvad (optional voice activity detection).

Pin librosa to a known stable version and ensure numba is compatible

Fix TTS error by updating save_pcm_to_wav to handle 16-bit audio samples correctly using NumPy tobytes()

Fix PermissionError by changing AUDIO_DIR to /tmp/audio_output for Hugging Face Spaces

Fix PermissionError by setting AUDIO_DIR to /app/audio_output for Hugging Face Spaces compatibility

Update audio handling: Save synthesized audio as WAV, return URL, set 10-min expiration with 5-min cleanup interval

Switch STT model to facebook/mms-1b-all with Whisper fallback

Fix STT error by removing return_attention_mask and address forced_decoder_ids warning

Fix MT error by removing clean_up_tokenization_spaces and address STT warnings

Update app.py

Add MT with NLLB-200-distilled-600M and TTS with MMS-TTS

Fix torch import error in translate-audio endpoint

Add STT functionality with openai/whisper-tiny

Add STT functionality with openai/whisper-tiny

Update README.md

Update app.py

Deploy minimal FastAPI app to test Hugging Face Spaces compatibility

Deploy minimal FastAPI app to test Hugging Face Spaces compatibility

Update app.py

Added pydantic for speech translation system

Update Dockerfile with proper caching and environment variables

Implement lazy loading of ML models to fix startup timeout on HF Spaces

Fix MT model loading: Revert to nllb-200-distilled-600M and add fallback

Further optimize memory usage: Switch to nllb-200-distilled-200M and add Uvicorn debug logging

Optimize app for Hugging Face Spaces: Switch to whisper-tiny, add health check, and improve model loading

Replaced the openai/whisper-tiny references with openai/whisper-small

Modified the _initialize_tts_model method to include the clean_up_tokenization_spaces parameter; Added logging configuration to configure the logging level for transformers in app.py

Modified the tokenization step to include clean_up_tokenization_spaces=True; Added clean_up_tokenization_spaces=True in the text_to_speech method; Added a print statement to confirm the TTS model is loaded

Removed the TRANSFORMERS_CACHE environment variable

Removed the TRANSFORMERS_CACHE environment variable and its associated debug print statement.

Added python-multipart

Update app.py

Set the Cache Directory Explicitly

Update requirements.txt

Create Dockerfile

Set sdk: docker

Removed SDK

Update requirements.txt

Removed flask, and flask-cors; Added fastapi, and uvicorn

Modified the code with API endpoints

Update app.py

Update README.md

Added flask, and flask-cors; Removed fastapi, uvicorn, and pydantic as dependencies

API integration for the Talklas pipeline

Added fastapi, uvicorn, and pydantic as dependencies