Guides · updated October 22, 2026

WAV file size

How large a WAV really is at 16-bit, 24-bit and 32-bit float, why a 2-hour file can pass 1 GB, and when RF64 is needed.

Tested: 22 October 2026 · 5 s and 60 min PCM fixtures · measured file sizes

Short answer

A WAV is sample rate × channels × bytes per sample × duration. At 44.1 kHz, 16-bit stereo it is about 10.1 MiB per minute, or roughly 600 MiB per hour. A standard RIFF WAV tops out at 4 GiB; beyond that the file needs RF64, which some older tools do not read.

The formula

bytes = sample rate × channels × (bit depth ÷ 8) × seconds

For one minute of 44.1 kHz, 16-bit stereo:

44,100 × 2 × 2 × 60 = 10,584,000 bytes

That is 10.6 MB (decimal) or 10.1 MiB (binary). The difference between MB and MiB is about 5%; both labels are common.

Common sizes

Format	Per minute	Per hour	2 hours
44.1 kHz, 16-bit, mono	5.3 MB	317 MB	635 MB
44.1 kHz, 16-bit, stereo	10.6 MB	635 MB	1.27 GB
48 kHz, 16-bit, stereo	11.5 MB	691 MB	1.38 GB
48 kHz, 24-bit, stereo	17.3 MB	1.04 GB	2.07 GB
96 kHz, 24-bit, stereo	34.6 MB	2.07 GB	4.15 GB
48 kHz, 32-bit float, stereo	23.0 MB	1.38 GB	2.76 GB

The 2-hour 96 kHz/24-bit file is already over 4 GB, which is why RF64 exists.

The 4 GB limit

The original RIFF/WAVE header uses 32-bit unsigned size fields. The practical maximum is just under 4 GiB. For longer or higher-resolution recordings, the industry uses RF64, an extension that adds 64-bit size fields. RF64 files are valid WAV files with a different header; not every old editor or hardware player reads them.

If you are writing a WAV from this tool and the estimated output approaches 4 GiB, the page warns before starting. For a huge file, consider:

A lower sample rate (48 or 44.1 kHz).
Mono instead of stereo.
16-bit instead of 32-bit float.
Trimming the audio.
A DAW or a tool that writes RF64.

Why WAV is large

WAV stores every sample, uncompressed. That is the point: a DAW can read it without decoding, and edits are sample-accurate. The size is the price of direct access. MP3, AAC and FLAC compress the samples; the first two are lossy, the third is lossless but still smaller than WAV.

What size does not tell you

A larger WAV is not a better master if the source was lossy.
A 32-bit float WAV is not “more detailed” than a 24-bit WAV; it has more headroom for processing.
A 96 kHz WAV is not automatically better than 48 kHz; it depends on the source and the capture chain.

Use the formula to plan storage, not to rank audio quality.

FAQ

How big is a 1-hour WAV?

At 44.1 kHz, 16-bit stereo: about 635 MB (605 MiB). At 48 kHz, 24-bit stereo: about 1.04 GB. At 96 kHz, 24-bit stereo: about 2.07 GB. The formula is sample rate × channels × bytes per sample × 3,600.

What is the 4 GB WAV limit?

The original RIFF format uses 32-bit size fields, so a standard WAV cannot exceed 4 GiB. RF64 extends the size fields and allows larger files. Some older editors and hardware players do not read RF64.

Is 32-bit float WAV worth the size?

For editing, yes: it keeps headroom and avoids clipping during processing. For delivery, 16-bit or 24-bit PCM is usually enough. A 32-bit float file is twice the size of 16-bit for the same sample rate and channel count.

Why is my WAV larger than the formula?

Metadata chunks, odd-sized chunks and RF64 headers add bytes. The audio data dominates, but a file with cover art or extensive metadata can be larger than the raw PCM calculation.

Does converting MP3 to WAV make the file lossless?

The WAV container is lossless, but the audio is still the decoded MP3. The file is larger; the detail is not restored. A WAV made from a 192 kbps MP3 is not a lossless master.