I've been working on a small tool recently, and it's finally live!
It's called TextFromAudio — a free online tool that converts audio into text.
The idea is pretty simple: upload an audio file, let the model process it, and get your transcript.
Why I built it
I wanted a simple audio-to-text tool that people could use without creating an account, dealing with subscriptions, or going through unnecessary steps.
Just upload your audio and get the text.
A little about the technology
I initially experimented with Sherpa-ONNX because I wanted to run speech recognition directly in the browser using WebAssembly.
It was an interesting approach, but after testing different options, I found that Whisper delivered noticeably better transcription results for my use case.
So I switched to Whisper.
I'm still experimenting with performance, transcription quality, and ways to make the experience smoother.
What's available?
- Free audio-to-text transcription
- Powered by Whisper
- Simple interface, no complicated setup
- No account required
It's still a small project, and there's plenty of room for improvement.
If you work with audio, podcasts, interviews, meetings, or just want to turn recordings into text, give it a try!
I'd love to hear your feedback, especially about transcription accuracy and performance.
Try it here: https://textfromaudio.io/
Top comments (0)