General Tech

Show HN: Overtone – Transcribe, correct and cut captions for any screen shape

Overtone is a tool that transcribes, corrects, and cuts captions for any screen shape, using a combination of machine learning models and a large language model to achieve a lower Word Error Rate (WER) compared to using a single model. It also features a rules-based cutter that can handle different screen shapes and aspect ratios, and supports various formats such as .overtone.json, .vtt, and .srt files. Overtone can be used offline or with API keys, and is open-sourced under the MIT license.

Read the full article at github.com →