🎤Speech to Text

Convert audio recordings to text using AI. Free, fast, and completely private.

Model Selection
Choose a model based on your speed and accuracy needs
Size
~150MB
Speed
Fast
Accuracy
Good
Record Audio
Click start to begin recording your voice
00:00
Or Upload Audio File
Drag and drop or choose an audio file from your computer
🎵

Drag and drop your audio file

Supports MP3, WAV, M4A, WebM, and other audio formats

How to Use

Option 1: Record Audio

  1. Select your preferred AI model (Tiny, Base, or Small)
  2. Click “Start Recording” to begin capturing audio from your microphone
  3. Speak clearly into your microphone
  4. Click “Stop Recording” when finished
  5. The AI model will download automatically (first time only, 150MB-970MB depending on model)
  6. Your transcription will appear within seconds

Option 2: Upload Audio File

  1. Drag and drop an audio file into the upload zone, or click “Choose File”
  2. Supported formats: MP3, WAV, M4A, WebM, and more
  3. The file will be processed locally in your browser
  4. Your transcription will appear automatically
Frequently Asked Questions

Is my audio data private?

Yes! Everything runs in your browser. Your audio never leaves your device.

What models are available?

We offer three Whisper models from OpenAI: Whisper Tiny (~150MB, fastest), Whisper Base (~290MB, balanced), and Whisper Small (~970MB, most accurate). All are optimized for browser use via Transformers.js. You can switch between models based on your needs for speed vs. accuracy.

Why is there a download on first use?

The AI model downloads once and is cached in your browser for future use. Size varies from 150MB to 970MB depending on which model you choose. Subsequent uses are instant as the model loads from cache.

Does it work offline?

After the first use, yes! The model is cached locally, so you can use it offline.

Coming Soon
🖼️

Image Resizer

Resize and optimize images

📄

PDF Converter

Convert files to/from PDF