Private, self-hosted audio and video to text
Your recordings stay on your own machine.
Txtify turns audio, video, and YouTube links into text and subtitles using Whisper (via
stable-ts),
with optional DeepL translation. It's free, open source, and self-hosted, so nothing gets
uploaded to someone else's cloud and no account is needed.
Transcribe audio and video in languages and translate transcriptions to languages. Export in multiple formats including . Files up to 1000MB and YouTube videos up to 15 minutes by default, both configurable when you self-host.
This site is a live demo to show how Txtify works. For real, private transcription, self-host it from the GitHub repository.
Transcribe audio and video in languages and translate transcriptions to languages. Export in multiple formats including . Files up to 1000MB and YouTube videos up to 15 minutes by default, both configurable when you self-host.
This site is a live demo to show how Txtify works. For real, private transcription, self-host it from the GitHub repository.
Enter the URL of the YouTube video you want to transcribe (default max 15 minutes, configurable).
Select the AI model to use for speech recognition.
Select the language spoken in the media.
Enable translation of the transcribed text.
Select the language to translate the transcribed text.