音视频 → 转写文稿 / 结构化总结:一条命令完成拆轨、人声分离、说话人分离转写与规范 docx;支持字幕、热词纠错、断点续跑与批量;默认离线,可选云端 ASR,纪要/总结由 AI Agent 按技能模板提炼。 | One-command AV → transcript & structured summary: demux, source separation, diarization, ASR, formatted docx; SRT, glossary fixes, resumable batches; offline-first with optional cloud ASR; Agent-driven minutes/summaries.
ffmpeg offline docx summarization speech-to-text transcription meeting-notes source-separation speaker-diarization video-to-text audio-to-text demucs llm pyannote funasr agent-skills claude-skills qwen3-asr av-processing
-
Updated
Nov 2, 2026 - Python