SRTWhisper

macOS app

A Mac app that turns video speech into subtitles.

I made this app because I needed it myself — I was always adding subtitles to videos, but every tool I found either wanted me to upload files to some server, or had an interface that was way more complicated than it needed to be. So I just built my own.

Under the hood, SRTWhisper uses OpenAI's Whisper model, but it runs entirely on your Mac — works even without internet. Toss a video in, and it extracts the audio track, runs speech recognition, and spits out a subtitle file with timecodes. Your files don't leave your machine.

Translation is also built in — supports local language models, the OpenAI API, and Ollama. Use whatever you have on hand. Output formats include SRT, VTT, and ASS. You can also hardburn the subtitles straight into the video.

Whisper isn't going to be 100% accurate every time — everyone knows that. So I built an editor into the app where you can play the video and fix the timing and text as you go, without switching to another tool.

Features

Speech recognition Whisper runs locally on your Mac. Choose from several model sizes (Tiny to Large-v3) depending on your needs and hardware. Works offline.
Translation Translate subtitles right after transcription. Supports local LLM, OpenAI GPT, and Ollama — covering English, Japanese, Korean, French, German, Spanish, and more.
Export SRT, WebVTT, and ASS formats. You can also burn subtitles directly into the video, so there's no extra post-production step.
Subtitle editor Edit text line by line, adjust start and end times, undo and redo. Watch the video while you edit, export when you're happy.
Interface languages English, Traditional Chinese, and Japanese. Switch anytime in Settings.

Privacy

Your videos and audio don't get uploaded anywhere. Speech recognition runs entirely on your Mac, there are no accounts, and the app doesn't collect usage data. The only time it connects to the internet is when you first download a Whisper model, or if you choose to use the OpenAI API for translation.