Why is WAV different from MP3 for transcription?

WAV is uncompressed — no lossy artifacts. Transcription accuracy is marginally higher, especially for quiet consonants like S, F, and T.

What bit depths are supported?

16-bit, 24-bit, and 32-bit (integer or float) PCM. Broadcast WAV (BWF) with BEXT metadata also handled.

Free tier: 2 GB and 5 minutes of audio. Registered users: 30 minutes/month. Pro: 10 hours/month. WAV is larger than MP3 — mind the size.

Metadata (timecode, description) is read for context but not modified. The output transcript is a separate text file.

Yes. Encrypted pipeline, auto-deleted within 24 hours, never used for AI training.

Yes. High-resolution WAV is downsampled server-side for transcription; the original file you uploaded is not modified.

Obtenez des transcriptions précises depuis des fichiers WAV non compressés. Idéal pour enregistrements pro, interviews studio et audio d’archive.

1
Upload your WAV file
Drop or browse for the WAV. We support PCM, 16/24/32-bit depth, mono or stereo, up to 192 kHz sample rate.
2
Select language
Pick from 99+ supported languages or auto-detect. Enable speaker separation for interviews and meetings.
3
Export the transcript
Download as TXT, DOCX, PDF, SRT, or VTT. WAV file deleted automatically within 24 hours.

All WAV formats
PCM WAV up to 192 kHz, 16/24/32-bit. Broadcast WAV (BWF) with metadata also supported.
Higher accuracy on lossless
WAV avoids MP3 compression artifacts. Transcription accuracy is typically 1-3 points higher than compressed formats.
99+ languages
Complete coverage including Mandarin, Japanese, Arabic, and rare languages.
Speaker separation
Identify and label up to 4+ distinct speakers in the audio.
Private processing
Encrypted storage, auto-delete after 24 hours, never used for AI training.

Studio interviews
Professional studios often record uncompressed WAV. Get clean transcripts without re-encoding artifacts.
Academic research
Field recordings and lab interviews are commonly archived as WAV. Transcribe for coding and publication.
Legal depositions
WAV is favored for legal recordings due to its lossless nature. Generate verbatim transcripts with speaker labels.

Conçu pour les créateurs, journalistes, étudiants et chercheurs du monde entier.

Why is WAV different from MP3 for transcription?: WAV is uncompressed — no lossy artifacts. Transcription accuracy is marginally higher, especially for quiet consonants like S, F, and T.
What bit depths are supported?: 16-bit, 24-bit, and 32-bit (integer or float) PCM. Broadcast WAV (BWF) with BEXT metadata also handled.
What is the size limit?: Free tier: 2 GB and 5 minutes of audio. Registered users: 30 minutes/month. Pro: 10 hours/month. WAV is larger than MP3 — mind the size.
Do you preserve BWF metadata?: Metadata (timecode, description) is read for context but not modified. The output transcript is a separate text file.
Are my WAV files private?: Yes. Encrypted pipeline, auto-deleted within 24 hours, never used for AI training.
Can I transcribe a 24-bit 96 kHz file?: Yes. High-resolution WAV is downsampled server-side for transcription; the original file you uploaded is not modified.