X-LANCE / SLAM-LLM

Speech, Language, Audio, Music Processing with Large Language Model

MIT License

579 stars 52 forks source link

Refactor, deepspeed support #49

Closed zzasdf closed 6 months ago

zzasdf commented 7 months ago

What does this PR do?

Simplify the config in example
Add decode script to the example asr_librispeech
Add user custom model
Merge the setup model with model_factory
move the setup_encoder and set_llm into model_factory
Move the inference function to slam_model_asr
Add deepspeed support
Move the checkpoint handler to utils