游雁 a23bdcd622 Merge branch 'vllm' - vLLM inference engine + streaming WebSocket service ...

Features:
- AutoModelVLLM: generic vLLM inference for all LLM-based ASR models
- serve_realtime_ws.py: streaming WebSocket ASR with VAD+SPK+hotwords
- DynamicStreamingVAD: dynamic silence threshold (default on for all models)
- Fun-ASR-Nano batch_size fix for VAD pipeline
- Comprehensive vLLM documentation (docs/vllm_guide.md)

2026-05-24 19:37:23 +08:00

..

auto

Merge branch 'vllm' - vLLM inference engine + streaming WebSocket service

2026-05-24 19:37:23 +08:00

bin

feat: auto-select model based on device (GPU=fun-asr-nano, CPU=sensevoice)

2026-05-24 06:03:08 +08:00

datasets

perf: optimize training pipeline for ~3% speedup

2026-05-23 20:56:53 +08:00

download

docs: add docstrings to 1950 functions across 276 files for API documentation

2026-05-21 07:52:52 +00:00

frontends

docs: add docstrings to 1950 functions across 276 files for API documentation

2026-05-21 07:52:52 +00:00

losses

docs: add docstrings to 1950 functions across 276 files for API documentation

2026-05-21 07:52:52 +00:00

metrics

docs: add docstrings to 1950 functions across 276 files for API documentation

2026-05-21 07:52:52 +00:00

models

Merge branch 'vllm' - vLLM inference engine + streaming WebSocket service

2026-05-24 19:37:23 +08:00

optimizers

docs: add docstrings to 1950 functions across 276 files for API documentation

2026-05-21 07:52:52 +00:00

schedulers

docs: add docstrings to 1950 functions across 276 files for API documentation

2026-05-21 07:52:52 +00:00

tokenizer

docs: add docstrings to 1950 functions across 276 files for API documentation

2026-05-21 07:52:52 +00:00

train_utils

feat: add bf16 mixed precision training support

2026-05-24 13:58:07 +08:00

utils

refactor: move DynamicStreamingVAD to funasr/models/fsmn_vad_streaming/

2026-05-24 18:26:08 +08:00

__init__.py

docs: add docstrings to 1950 functions across 276 files for API documentation

2026-05-21 07:52:52 +00:00

register.py

docs: add docstrings to 1950 functions across 276 files for API documentation

2026-05-21 07:52:52 +00:00

version.txt

release: v1.3.3 - add funasr-server CLI + OpenAI-compatible API + MCP server

2026-05-24 02:17:58 +08:00