Convert MP4 to WAV
Pull the audio out of an MP4 as an uncompressed WAV. The picture is discarded and the sound is decoded to raw samples, which is what audio editors, transcription engines and analysis tools generally want. It will not sound better than the MP4 did — nothing recovers what an encoder discarded — but it will open in software that refuses compressed audio.
Use this without the search next time. Prathom Workbench puts Prathom's tools in your toolbar.
Add to Chrome — freeDrop your MP4 file here, or click to browse
Up to 50 MB. Deleted automatically after 30 minutes.
What it does
- The video stream is discarded, not re-encoded
- Uncompressed 16-bit PCM output
- For editors and transcription tools that require WAV
- No watermark and no sign-up
How to use MP4 to WAV
- 1
Add the MP4
Drop the file in or choose it from your computer. The 50 MB limit applies to the upload, so long recordings may need trimming.
- 2
Convert
The video is dropped and the audio is decoded to 16-bit PCM and written as a WAV.
- 3
Import it
The WAV loads into every audio editor and into tools that reject compressed formats.
The picture is not converted, it is left behind
Extracting audio is not really a conversion of the video. The picture is not decoded, not re-encoded and not written anywhere — it is simply not carried into the output. That is why a large video becomes an audio file quickly, and why the result is smaller than the video despite WAV being an uncompressed format.
What is actually happening is that one stream out of two is decoded, and its samples are written out at full precision.
How it works
The audio track — AAC, in nearly every MP4 — is decoded to 16-bit PCM samples, and those samples are written into a WAV container with no further processing. There is no encoder on the output side. WAV stores samples as they are, which is both why every piece of audio software reads it and why the file is large.
Why so much software insists on uncompressed audio
It looks like obstinacy from the outside and it is usually caution. Reading a WAV needs no decoder, so there is no library to depend on, no format to license, and nothing between the file and the samples.
For anything doing measurement rather than playback, that last point is the real argument. Lossy encoders shape audio around a model of what a person will notice. A transcription engine, a pitch tracker or a forensic analysis tool is not a person, and starting from uncompressed samples removes the encoder's judgments from the result.
Plan for the size
Uncompressed stereo at CD quality is about ten megabytes a minute, and it is that whether the passage is a full mix or silence, because every sample is stored at full size regardless.
An hour of recorded meeting is therefore around six hundred megabytes, and a day of interviews runs into several gigabytes. If the tool at the other end accepts mono — and for people talking into a laptop microphone it generally should, since there is no stereo image worth preserving — converting to mono halves the figure at no meaningful cost.
What it cannot do
It cannot improve the recording. If the audio in the MP4 was encoded at a low bitrate by conferencing software, the WAV is a large, exact copy of that. Decoding is complete and faithful; it is not restorative, and no format conversion has ever been.
Examples
Audio from a recorded meeting for transcription
Transcription and speech-analysis engines used in research, legal and medical work frequently accept WAV only, and a video file is what the meeting software produced.
A soundtrack for an audio editor
Older editors and some scientific tools import uncompressed audio and nothing else. Extracting it directly avoids a separate decoding step in the middle.
Frequently asked questions
Will the audio sound better as a WAV?
No. The MP4's audio is almost always AAC, which is lossy — the encoder permanently removed what it judged inaudible before the file existed. Decoding reproduces exactly what the MP4 contains at full precision, which is what it already sounded like. The file becomes larger and more widely accepted, not better.
Why do transcription tools want WAV rather than MP4?
Because uncompressed audio removes variables. There is no decoder dependency, no codec licensing, and no possibility that the encoder's own shaping affects a measurement. That last point carries real weight for speech analysis: a lossy encoder is built around a model of human hearing, and a tool measuring pauses, pitch or formants is not a human ear.
How large will the WAV be?
Around ten megabytes a minute for stereo at CD quality, regardless of content. An hour-long meeting recording becomes roughly six hundred megabytes. If your tool accepts mono — and for a recording of people talking it usually should — converting to mono halves that, and loses nothing that matters for speech.
Should I keep the MP4?
Yes. It is far smaller, it holds the same audio, and it also holds the video. Treat the WAV as a working file for the session that needed it and delete it afterwards; you can always extract again.