word-level-timestamps 话题在 GitHub 上没有填写简介，这里按 Star 数汇总最相关的热门仓库。
Lingua 是一个面向语言学习的播放器。它把音频、字幕、逐词时间轴、注音、转写、词性、译文和词汇讲解放在同一条学习流程。支持日语、英语、西班牙语、法语、德语和韩语。Lingua is a media player designed for language learners. It integrates audio, subtitles, word-by-word timelines, phonetic symbols, transcription, parts of speech, translations, and vocabulary explanations into a single learning flow.
LyricsEditor - 免费开源的高级歌词与字幕工具。一款专业级、基于浏览器的逐词歌词与字幕编辑器。专为精确、高效和广泛格式兼容性而设计。
The same caption renderer in your live preview and in your FFmpeg pipeline. Pixel-identical output, no headless browser.
智能体式视频字幕质量检查与字幕渲染 CLI：上下文 ASR 纠正、词级强制对齐、活动词 ASS 字幕和校验帧。
Local-first AI subtitle generator for anime, film & clips. Vocal separation + song-skip keep BGM, OP/ED and insert songs out of the ASR, so Qwen3-ASR does not hallucinate on music. Word-level forced alignment, speaker diarization, edit-and-resync, SRT/VTT/ASS. CJK-aware (中文 / 日本語 / English).
MMS Forced alignment package for JavaScript, TypeScript and Node.js: word-level timestamps from audio + transcript for captions, subtitles and karaoke. Meta MMS (wav2vec2) via ONNX Runtime with a CTC Viterbi decoder. Hindi, Hinglish and 20 Indic languages. No Python.
Auto-generate word-level subtitles for Telugu & Indian English videos using a fine-tuned Whisper model. Extracts audio via FFmpeg, transcribes with strict word-level timestamps, and outputs .srt files ready for CapCut, VN Editor, and Adobe Premiere Pro.
The missing editor for WhisperX using word-level timestamps. Fix JSON timing precisely captions and viral videos before rendering.