Audio to Text (AI)
Transcribe any audio with AI, 100% in your browser. Your audio is never uploaded anywhere.
Drag your audio here or click to select
MP3, WAV, OGG, FLAC, M4A, AAC, Opus, MP4 · no artificial size limit
Built by
Miguel Ángel Colorado Marin (MACM)
Built by
Miguel Ángel Colorado Marin (MACM)
Full-Stack Developer · Guadalajara, España
I develop web apps, digital tools and full projects — from design to deployment.
The audio to text transcriber uses Whisper, OpenAI's open-source speech recognition model, running entirely inside your browser with WebGPU or WebAssembly — your audio is never uploaded to any server. Before transcribing, a voice-detection model (Silero VAD) finds the spans where someone is actually speaking and discards silence, so the result doesn't break down or repeat itself on long recordings with pauses. Upload a recording, a voice note, a podcast or any audio file (MP3, WAV, OGG, FLAC, M4A, AAC, Opus or MP4), choose the model (fast, precise or max quality) and the language, and within seconds you get the transcribed text, ready to edit, copy or download as .txt. The first time you use it the models download and stay cached in your browser, so following transcriptions are much faster. It's the same technology paid services use, but running on your own device, at no cost, with no artificial size limit and your audio never leaving your computer or phone — the only real limit is what your own device can process.
Features
- ✓OpenAI's Whisper model (open source)
- ✓Automatic language detection
- ✓No minute limits
- ✓Edit, copy and download .txt
- ✓100% in browser, your audio is never uploaded
How to transcribe audio to text?
- 1
Upload your audio
Drag the file or click to select it. No artificial size limit.
- 2
Choose the model and language
Fast, precise or max quality, and the audio's language (or auto-detect).
- 3
Tap Transcribe
The first time it downloads the AI model; after that it stays cached.
- 4
Edit, copy or download
The text is directly editable. Copy it or download it as .txt.
Frequently asked questions
Is my audio uploaded to a server?
No. The Whisper model runs inside your browser with WebGPU or WebAssembly. The audio never leaves your device.
What's the difference between the three models?
Fast (~75 MB) transcribes sooner but with slightly less accuracy. Precise (~150 MB) is the recommended balance. Max quality (~1 GB, Whisper large-v3-turbo) takes longer and is a bigger download but gets more right, especially with accents, noisy audio or specific vocabulary.
Is there a size limit?
There's no artificial limit — the real limit is what your device can decode and process. Very large files just take longer.
What formats and languages does it support?
MP3, WAV, OGG, FLAC, M4A, AAC, Opus and MP4 as input. Whisper is multilingual: choose Spanish, English, French, German, Italian, Portuguese, or let it auto-detect.
Is it always 100% accurate?
No, like any transcription AI — though the voice-detection step upfront avoids the most common failure (looping/repeating on long recordings with pauses). It can still struggle with strong accents, heavy background noise or very technical vocabulary — always review the result, and if something looks off try "Max quality" or force the language instead of leaving it on auto.
Related tools
Embed Audio to Text (AI) on your site
Add Audio to Text (AI) to any web page with a simple iframe. Free, with attribution to miguelacm.es.
<iframe
src="https://miguelacm.es/embed/audio-to-text"
width="100%"
height="700"
frameborder="0"
title="Audio to Text (AI) — miguelacm.es"
></iframe>