v1.3.10

datalab-to/markerv1.3.10Jun 19, 2026by LauraGPT

AI Summary

This release is for modelscope/FunASR repository, not datalab-to/marker. Introduces agent-friendly CLI, batched VAD-segment decoding, and various fixes for Fun-ASR-Nano and WebSocket server.

Key Highlights

New agent-friendly CLI with structured output support
Fun-ASR-Nano now supports batched VAD-segment decoding (~1.75× faster)
WebSocket 2-pass server with sentence-level timestamps
Multiple bug fixes for fp16/bf16 inference and CLI
New vLLM server flags for VAD and speaker models

New Features

Agent-friendly CLI with JSON output format
Batched VAD-segment decoding for Fun-ASR-Nano
WebSocket 2-pass server with sentence timestamps
New --vad-model and --spk-model flags for vLLM server
Improved CLI timestamp handling and model ID support

Full Release Notes

## FunASR v1.3.10

### New features
- **Agent-friendly CLI**: `funasr audio.wav --output-format json` for structured output
- **Fun-ASR-Nano**: batched VAD-segment decoding (~1.75× faster) (#2979)
- **WebSocket 2-pass server**: sentence-level timestamps
- **serve_vllm.py**: new `--vad-model` / `--spk-model` flags

### Fixes
- Fun-ASR-Nano: bf16/fp16 inference no longer crashes; warn on degraded fp16 (#2980)
- Fun-ASR-Nano vLLM: fix CUDA crash from `repetition_penalty`
- CLI: valid SRT timestamps + correct JSON durations (#2982); use `sentence_info` text (#2983); correct model id `Fun-ASR-Nano-2512` (#2984)
- Clearer error for missing audio path (#2981); respect explicit VAD silence threshold; handle `None` encoder/scheduler configs

### Docs
- New CLI reference; clearer vLLM install guidance

**Full changelog**: https://github.com/modelscope/FunASR/compare/v1.3.9...v1.3.10