Audio to PDF Transcriber
AI-powered · Word-accurate · Private
Model & Language Fast · Auto-detect
Drop an audio file or tap to upload
MP3 · WAV · M4A · OGG · FLAC · WEBM · AAC
Up to 24 MB · your file is streamed securely to the AI Gateway
AI-Powered Accuracy
State-of-the-art speech recognition trained on 99+ languages — even messy voice notes come out clean.
Private by Default
Audio streams straight to the transcription model and is never stored on our servers.
Edit Before Export
The transcript is editable — fix names, punctuation, and speaker labels before you save the PDF.
How it works
Upload or record
Drop an audio file, or record straight from your mic in-browser
AI transcribes
The AI Gateway runs a modern ASR model and returns your transcript
Edit the text
Clean up names, punctuation and paragraph breaks inline
Download PDF
One tap — get a formatted PDF or plain .txt file
Frequently Asked Questions
Which audio formats are supported?
MP3, WAV, M4A, OGG, FLAC, WebM and AAC. Browser recordings (WebM / MP4) are handled automatically.
Is there a file size limit?
Yes — up to 24 MB per file. For longer recordings, split them first with an audio editor and merge the transcripts.
Which languages does it support?
99+ languages including English, Spanish, Hindi, French, German, Japanese, Chinese and Arabic. Set Auto-detect to let the model choose.
Is my audio kept anywhere?
No. Files are streamed to the transcription model and are not persisted by PDFNinja.
Does it work with multiple speakers?
The transcript is a single flowing text — speaker diarization is planned. For now, you can add speaker labels manually before exporting.