{"version":1,"type":"story","url":"https://digestai.news/story/overtone-releases-tool-to-cut-captions-for-any-screen-shape","json":"https://digestai.news/story/overtone-releases-tool-to-cut-captions-for-any-screen-shape.json","markdown":"https://digestai.news/story/overtone-releases-tool-to-cut-captions-for-any-screen-shape.md","slug":"overtone-releases-tool-to-cut-captions-for-any-screen-shape","headline":"Overtone releases tool to cut captions for any screen shape","summary":"Overtone is a new open-source tool that transcribes, corrects, and formats video captions for multiple screen shapes from a single file. It stores each spoken word once with its timestamp and speaker, then generates separate caption cuts for different aspect ratios, such as 16:9 for desktop and 9:16 for social media. The tool uses a CLI to process subtitles, audio, or video, and includes a video.js plugin for playback.\n\nThe system combines multiple speech engines, including OpenAI and Azure AI Speech, to improve accuracy. It uses an LLM to correct misheard words and names based on context, while ensuring timestamps remain accurate by relying on the speech engines for timing. Benchmarks show Overtone achieves a 5.3% word error rate, outperforming individual engines like gpt-transcribe (6.2%) and Azure Speech (7.7%). It also significantly reduces line breaks ending on weak words, with only 0.3% of breaks ending on words like \"the\" or \"of\", compared to 28.8% for standard line-filling methods.\n\nThe tool is designed for offline use without API keys for recutting existing subtitles, though full transcription requires provider keys. It supports languages including English, Dutch, German, French, Spanish, Italian, and Portuguese, with specific rules for line breaking and number handling. The project is public, with full method details and caveats available in its documentation.","keyPoints":["Overtone cuts captions for 16:9 and 9:16 screens from one file, storing words once.","Benchmarks show 5.3% word error rate, better than gpt-transcribe's 6.2% and Azure's 7.7%.","Only 0.3% of line breaks end on weak words, compared to 28.8% for standard methods."],"whyItMatters":"Solves a common pain point for video creators needing captions for both desktop and mobile. Improves accessibility and viewer experience by ensuring clean, readable captions across all devices without manual reformatting.","category":{"slug":"agents","name":"Agents & Tools","url":"https://digestai.news/category/agents"},"entities":{"companies":["OpenAI","Azure AI","ElevenLabs","Blender Foundation"],"models":["gpt-transcribe","whisper.cpp","Ollama"],"people":[]},"firstPublishedAt":"2026-10-29T17:48:17Z","updatedAt":"2026-10-29T17:48:17Z","sourceCount":1,"hasPrimarySource":true,"sources":[{"outlet":"github.com","title":"Show HN: Overtone – Transcribe, correct and cut captions for any screen shape","url":"https://github.com/lonthq/overtone-captions","publishedAt":"2026-10-29T17:48:17Z","type":"primary","primary":true,"lead":true}],"sourceNotes":null,"discussions":[{"site":"Hacker News","url":"https://news.ycombinator.com/item?id=50293115","points":41}],"thread":null,"cite":{"text":"Digest AI, \"Overtone releases tool to cut captions for any screen shape\", 29 October 2026, https://digestai.news/story/overtone-releases-tool-to-cut-captions-for-any-screen-shape","publisher":"Digest AI","title":"Overtone releases tool to cut captions for any screen shape","datePublished":"2026-10-29T17:48:17Z","url":"https://digestai.news/story/overtone-releases-tool-to-cut-captions-for-any-screen-shape"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}