whereismyai
Overtone Captions
Overtone Captions is an open-source CLI tool that stores spoken words once and cuts subtitles separately for different screen shapes.
Overtone Captions is an open-source caption management tool that generates separate caption cuts for different screen shapes from a single data file. It stores each spoken word once with its timestamp and speaker, then cuts captions for 16:9, 9:16, or custom aspect ratios. The tool uses multiple speech and text engines including OpenAI, Azure AI Speech, and ElevenLabs Scribe, combined with an LLM editor for correction and a versioned rule set for line breaks. It handles offline execution using whisper.cpp and Ollama, or runs via a command line interface. The package includes a video.js plugin for playback and supports multiple language rules, exporting standard WebVTT, SRT, and text formats.
Overtone Captions at a glance
Key features
- Stores each spoken word once with time and speaker
- Cuts captions separately for 16:9, 9:16, and custom aspect ratios
- Uses multiple speech engines for transcription and timing
- Corrects misheard words and names with an LLM editor
- Follows reading rules to avoid weak line endings
- Exports WebVTT, SRT, and text files per cut
- Includes a video.js plugin with style packs
Use it for
- Re-cutting existing subtitles for vertical and horizontal video
- Transcribing and generating multi-aspect captions from audio
- Displaying responsive captions in video.js web players
Alternatives to Overtone Captions
- AuK — It handles zero-shot TTS, editing, and enhancement via natural language.
- MOSS-TTS — Delivers zero-shot voice cloning and fine-grained pronunciation control.
- Applio — Runs locally with open-source code and no telemetry.
- GPT-SoVITS — It allows high-fidelity voice cloning with as little as one minute of data.
- Whishper — Transcribe and edit subtitles 100% locally with an integrated web UI.
- VoxCPM — Provides tokenizer-free multilingual speech with controllable voice creativity.
- Dub Anything — Runs locally on your PC and uses an AI coding agent to direct the dub line by line.
- Voicebox — Perform voice cloning and generation entirely on local hardware.
- Faster Whisper GUI — Provides a visual interface for faster-whisper, whisperX, and Demucs models.
- Async — Edit videos by chatting with an AI trained on professional editing skills.
- IndexTTS — Supports zero-shot voice cloning with multi-language emotion and speed control.
- MisoTTS — Generates high-quality conversational speech using an RVQ Transformer architecture.
All Overtone Captions alternatives