Drop any voice recording and get a clean, accurate transcript in seconds. No setup. No API keys. No guesswork.
// what you get
Whisper-powered accuracy
OpenAI Whisper runs under the hood. Choose from Tiny to Large depending on speed vs. accuracy needs.
Auto language detection
Whisper identifies the spoken language automatically, or you can pin a specific one. 99+ languages supported.
Copy, download, share
One-tap copy to clipboard, or save as a clean .txt file named after your original recording.
Secure proxy, no leaks
Your audio travels through a Vercel edge proxy. No third-party analytics. No ad trackers.
Mobile-first design
Built for phones from the ground up. Tap to upload, choose settings, and read the result.
Live server health
The status pill pings the backend every 30 seconds so you always know before you upload.
// how it works
Upload your audio
Tap the upload zone and pick your file, or drag and drop it. Supports .m4a, .wav, .mp3, .ogg, .flac, and .aac.
Choose your settings
Pin a language or let Whisper auto-detect. Select a model tier: Tiny for speed, Large for accuracy.
Get your transcript
Results in under 5 seconds for short clips. Copy or download as .txt. Word count and language metadata included.
// supported formats
// languages include
EnglishHindiSpanishFrenchGermanArabicJapaneseChinesePortugueseKoreanRussian+ 88 moreCreate a free account or jump straight into the app.