How to compress wav to mp3 without uploading — free browser tool
- Step 1
- Step 2Drop the oversized WAV — Drag the large
.wavin. It's read into browser memory only — no upload, so there's no slow transfer of the big file before you can act. - Step 3Pick a bitrate to set compression strength — Lower bitrate = smaller file. For aggressive shrink choose
128 kbps · podcastor64 kbps · voice; to compress gently choose256kor320k. This is the only control and it directly maps to FFmpeg-b:a. - Step 4Compress — Run it.
libmp3lamere-encodes the WAV to MP3 in the browser. The size drop is immediate and predictable from the bitrate. - Step 5Download and check the size — Save the MP3 and confirm it's under your target. If it's still too big, re-run at a lower bitrate, or switch to a size-targeting tool.
- Step 6If you need an exact size cap — When the requirement is 'under N MB' rather than 'good enough quality', use audio-compressor (target megabytes), or platform presets like discord-fit and whatsapp-fit.
How much each bitrate compresses a WAV
Approximate output size and compression ratio for a 100 MB stereo WAV (about 10 minutes of CD-quality audio). Ratios are roughly constant per minute.
| Bitrate | Output for a 100 MB WAV | ~Compression ratio | Quality note |
|---|---|---|---|
| 320 kbps | ~24 MB | ~4x | Near-transparent; gentle compression |
| 256 kbps | ~19 MB | ~5x | Transparent to most ears |
| 192 kbps (default) | ~14 MB | ~7x | Standard distribution quality |
| 128 kbps | ~9 MB | ~11x | Good for voice + casual music; noticeable on dense mixes |
| 64 kbps | ~5 MB | ~20x | Voice only; not for music |
Bitrate target vs size target — pick the right tool
This converter compresses by bitrate. When you have a hard megabyte ceiling, a size-targeting sibling works backward to fit it.
| You want | Tool | How it works |
|---|---|---|
| A specific quality (bitrate) | wav-to-mp3 (this tool) | You pick 64k-320k; size follows |
| Under a custom MB target | audio-compressor | You set target megabytes; it solves for bitrate |
| Under Discord's limit | discord-fit | Targets a 25 MB ceiling automatically |
| Under WhatsApp's limit | whatsapp-fit | Targets a 16 MB ceiling automatically |
| Smaller WAV that stays lossless | wav-to-flac | Lossless FLAC, ~40-60% of WAV size, no quality loss |
Cookbook
Real 'this WAV is too big' situations and the bitrate that solves each. The FFmpeg shown is what the in-browser engine runs.
300 MB session WAV down to an emailable file
A 30-minute stereo WAV at ~300 MB won't attach to email. At 192 kbps it lands around 41 MB; at 128 kbps around 27 MB — under most email caps.
Source: session.wav (44.1 kHz / 16-bit / stereo, ~300 MB)
Option: 128 kbps · podcast
Engine runs:
ffmpeg -i session.wav -c:a libmp3lame -b:a 128k \
-map_metadata 0 -id3v2_version 3 session.mp3
Result: session.mp3 (~27 MB) — ~11x smallerVoice recording crushed for cloud storage
An interview WAV that's all speech can drop to 64 kbps with no meaningful loss, for maximum shrink on a storage quota.
Source: interview_raw.wav (mono, 50 min, ~250 MB)
Option: 64 kbps · voice
Engine runs:
ffmpeg -i interview_raw.wav -c:a libmp3lame -b:a 64k \
-map_metadata 0 -id3v2_version 3 interview_raw.mp3
Result: interview_raw.mp3 (~24 MB) — ~10x smallerGentle compression that keeps music quality
When you want to free space but keep the music sounding good, 256 kbps cuts about 5x while staying transparent to most listeners.
Source: live_set.wav (~180 MB)
Option: 256 kbps
Result: live_set.mp3 (~36 MB) — ~5x smaller,
near-transparent
For an exact ceiling instead, see audio-compressor
(/audio-tools/audio-compressor).Still too big? Drop a bitrate step
If 192 kbps didn't get you under the limit, re-run at 128. There's no penalty for converting again from the same WAV — you're always encoding from the lossless source, not stacking MP3 passes.
First try: 192 kbps -> 38 MB (over a 25 MB cap) Re-run: 128 kbps -> 25 MB (just fits) Always encode from the original WAV, not from a previously-made MP3, to avoid stacking lossy passes.
When FLAC beats MP3 for shrinking
If you need the file smaller but cannot lose any quality (archival), FLAC compresses losslessly to roughly half the WAV size. MP3 goes smaller but is lossy. Choose by whether 'lossless' is a requirement.
100 MB WAV -> wav-to-flac (/audio-tools/wav-to-flac) -> ~55 MB, lossless wav-to-mp3 @ 192k (this tool) -> ~14 MB, lossy FLAC: smaller-than-WAV with zero loss. MP3: far smaller, slight loss.
Edge cases and what actually happens
WAV is over your tier's size limit, so you can't even load it to compress
BlockedIronically the file you want to shrink may exceed the input cap: Free 50 MB, Pro 200 MB, Pro-media/Developer 100 GB. If a giant WAV won't load, upgrade your tier, or split it into parts first and compress each — the output MP3 will be small, but the input must fit the limit.
Output still too big at your chosen bitrate
ExpectedSize is bitrate times duration — a long file is big even at 128 kbps. Drop to 64 kbps (voice only), or use audio-compressor to target an exact megabyte ceiling. There's no single magic setting that fits arbitrary length under a fixed cap except a size-targeting tool.
You wanted to target exactly N megabytes
Wrong toolThis converter compresses by bitrate, not by target size. For 'must be under 25 MB', use audio-compressor; for platform caps use discord-fit or whatsapp-fit. They solve for the bitrate that fits.
Heavy compression made music sound bad
Quality caution64 kbps is fine for speech but produces audible artifacts on music — smeared cymbals, swirly high frequencies. If a low-bitrate music file sounds rough, step back up to 128 or 192 kbps and accept a larger file, or compress losslessly with wav-to-flac.
Compressing a lossy-origin WAV
Quality cautionIf the WAV came from decoding an MP3/AAC, compressing it back to MP3 adds a second lossy pass and can sound noticeably worse at low bitrates. Keep the original compressed file or a true lossless master where you can.
Expected sample rate to drop for more compression
By designThe converter preserves sample rate — it compresses via the codec, not by downsampling. To also reduce sample rate (e.g. 44.1 kHz to 22.05 kHz for voice) run sample-rate-converter first, then encode at a low bitrate for extra shrink.
Expected mono downmix to halve the size
By designThis tool keeps the channel count. A stereo file stays stereo. To halve data via mono, downmix before converting (e.g. with a WAV-target conversion tool that exposes channels), then compress to a low bitrate.
Long file compresses slowly
ExpectedWebAssembly runs on the CPU, so a 50-minute WAV takes longer than a short clip. It still runs locally with no upload; allow time on long inputs and older devices.
Tags lost after compression
PreservedTags and art are copied to ID3v2.3 during the encode. If a player shows them missing, it may be caching; verify with id3-editor. They are written regardless of bitrate.
Engine won't load before first cache
Load errorFFmpeg WebAssembly fetches on first use. If you're offline or the CDN is blocked, compression can't start. Reconnect and reload; after the first run it works offline.
Frequently asked questions
How much will compressing my WAV to MP3 save?
Roughly 4x at 320 kbps, 7x at 192 kbps, 11x at 128 kbps, and about 20x at 64 kbps. A 100 MB WAV becomes ~24 MB, ~14 MB, ~9 MB, or ~5 MB respectively. Lower bitrate compresses harder at the cost of quality.
Does compressing upload my big WAV first?
No — and that's the point. The whole job runs in your browser via FFmpeg WebAssembly, so there's no slow upload of the oversized file. The WAV is read locally, compressed locally, and the small MP3 is downloaded.
Which bitrate compresses the most?
64 kbps produces the smallest file (about 20x smaller than the WAV) and is the strongest compression here, but it's only suitable for voice. For music, 128 kbps is the practical floor before artifacts become obvious.
I need it under a specific size — how?
Use a size-targeting tool: audio-compressor lets you set a megabyte target; discord-fit and whatsapp-fit auto-fit those platforms' caps. This wav-to-mp3 tool compresses by bitrate, so size follows your bitrate choice rather than the reverse.
Can I compress losslessly instead?
Yes, with FLAC. wav-to-flac compresses to roughly half the WAV size with zero quality loss. It won't shrink as much as MP3, but nothing is lost — ideal for archival.
Will compression reduce my sample rate or channels?
No. It preserves both and saves size purely through codec compression. To also cut sample rate or fold to mono for extra shrink, use sample-rate-converter and a channel/downmix step before converting.
My WAV is too big to even load — what now?
The input must fit your tier's cap (Free 50 MB, Pro 200 MB, Pro-media/Developer 100 GB). If it doesn't, split the WAV into smaller parts first and compress each, or upgrade. The output MP3 will be small either way, but the source has to fit the limit.
Is the quality loss noticeable?
At 192-320 kbps, rarely for most listeners. At 128 kbps it's audible on dense music but fine for voice and casual listening. At 64 kbps music suffers but speech is acceptable. Pick the highest bitrate that still meets your size goal.
Can I re-compress an MP3 to make it smaller?
You'd want bitrate-changer for an existing MP3, not this WAV tool. Re-encoding an MP3 to a lower bitrate stacks lossy passes, so always compress from the original WAV if you still have it.
Does it remove silence to save space?
No — this tool only changes the bitrate. To trim leading/trailing or mid-file silence (which does cut size), use silence-stripper first, then convert the shorter WAV to MP3.
How long does compressing a big WAV take?
Seconds for a few minutes of audio; longer for hour-long files since WebAssembly runs on your CPU. There's no upload wait, so total time is just local encode time, which scales with duration.
Is there a watermark or quality cap on the free tier?
No watermark and no forced quality cap — you can pick any of the five bitrates including 320. The free tier limits input size (50 MB) and duration (30 min) per file, not output quality.
Privacy first
Every JAD Audio tool runs entirely in your browser via FFmpeg (WebAssembly) and RNNoise. Your audio files never leave your device — verified by zero outbound network requests during processing.