Open-source LLM-based ASR model family for Chinese, dialect, accent, and multilingual speech, with FunASR, vLLM, streaming, and llama.cpp runtimes.
audio pytorch speech-recognition speech-to-text transcription asr speaker-diarization on-device chinese-dialects llama-cpp real-time-asr gguf funasr audio-language-model multilingual-asr fun-asr whisper-alternative 31-languages llm-asr fun-asr-nano
-
Updated
Jul 14, 2026 - C