DroolingPanda/teachingAssistant

fix variable name bug

fix(stt): change default model from whisper-large to whisper-medium

fix: remove hard-coded ASR model list and make ASR model optional

chore: remove unused dependencies and update model selection in Whisper provider

fix(stt): handle whisper-large model name as alias for whisper provider

fix(stt): improve provider selection and error handling in STT service

refactor: remove parakeet ASR provider and update all references to Whisper only

perf(stt): optimize device handling and use model.generate for inference

fix(logging): use exc_info=True instead of exception=error in STT provider

fix(config): update supported ASR models list to reflect actual availability

feat: remove unused language options and update supported languages list

feat: restrict supported languages to English and Chinese only

clean up

remove tests

feat(logging): add detailed logging for audio upload and configuration handling

Update README.md

Update README.md

Switch to use gradio

chore: update dependencies and replace NeMo with HF transformers for Parakeet STT provider

chore(deps): remove version pinning in requirements.txt

feat: add pre-requirements.txt and remove Cython from requirements.txt

chore(deps): add Cython to requirements

feat: replace legacy TTS providers with Chatterbox as the single, default provider

fix

fix

fix

remove all tts providers

try fixing gr.blocks issue

try fixing build issue

update requirements, just to get chatterbox working

add chatterbox

remove legacy impl

upgrade gradio

Remove Streamlit dependency and simplify Dia TTS dependency handling

use Gradio

update descript-audio-codec version

add descript-audio-codec

attempt dia fix

add more logs

set parakeet model to default asr

change parakeet to be default asr

fix dia tts

set parakeet as default stt model

attempt fix

attempt to fix cosyvoice2 tts

fix error

fix error code

fix validation error

fix logging issue

log