Audio Bitrate Reducer: Shrink Files Without Guessing

The audio bitrate reducer re-encodes an audio file at a lower bitrate and reports the real before-and-after sizes measured on disk. It reads the source with ffprobe, encodes with ffmpeg, and shows you what actually happened — including the case where the output came out larger.

That last point matters. Many "compress audio" pages show an invented saving percentage. This one compares two files. If you ask for a bitrate above what the source already uses, the result can grow, and the tool says so instead of hiding it.

#What you can change

• Output format — MP3, AAC (.m4a), Opus, FLAC or WAV.
• Bitrate — auto, or manual from 8 to 320 kbps.
• Channels — keep, mono or stereo.
• Sample rate — keep, or 8,000 to 48,000 Hz. It never upsamples.
• Input formats — MP3, WAV, M4A, AAC, OGG, OGA, Opus, FLAC, WMA, WebM, AIFF and AMR, up to 50 MB.

The output is audio only. If the input was a video file with an audio track, or an M4A with embedded cover art, those extra streams are dropped. That is often a meaningful part of the size reduction, and the notes say when it happened.

#Automatic versus manual bitrate

Auto mode looks at the source and aims below it: about 70% of the source bitrate for stereo, 60% for mono, with a 160 kbps ceiling for mono. For MP3, the target snaps to a standard tier — 64, 96, 112, 128, 160, 192, 256 or 320 kbps — so the result matches something you would recognise from an export menu.

Manual mode applies the rate you choose, within 8 to 320 kbps. Use it when you know the destination: 64 kbps mono for a voice note, 128 kbps for a podcast, 192 kbps for music that will be listened to on good headphones.

#How to reduce an audio file's bitrate

1. Open the audio bitrate reducer.
2. Choose an audio file within the 50 MB limit.
3. Pick the output format. MP3 is the most compatible; Opus is more efficient at low bitrates; AAC works well in Apple ecosystems.
4. Leave bitrate on auto, or choose a manual target.
5. Switch to mono if the file is speech and you do not need stereo.
6. Leave the sample rate on keep unless you specifically need to lower it.
7. Press Run and compare the reported sizes before downloading.

The original is not modified. The download is named after the source with a -reduced suffix, so it will not overwrite the file in your downloads folder.

#Why the output sometimes gets bigger

Re-encoding at a higher bitrate than the source cannot add detail that was never there, but it can add bytes. The encoder still has to store the signal at the requested rate, and container overhead differs. The tool reports a negative saving as a loss and explains it.

A file can also grow when you convert between lossy formats at the same nominal bitrate. Decoding MP3 and re-encoding to AAC is a second generation of lossy compression; the result may be larger and slightly worse. If compatibility is not the issue, keep the original format.

#Lossless formats behave differently

FLAC and WAV are lossless. There is no bitrate to target, and the encoder ignores the bitrate setting. Choosing FLAC makes sense when the source is an uncompressed WAV or AIFF that has never been compressed: FLAC can cut that file roughly in half without losing anything.

Choosing WAV makes sense when you need an uncompressed file for editing, not when you want a smaller one. A WAV is usually larger than the MP3 you started with. The tool reports the real size either way and marks whether a target bitrate was actually applied.

#Common problems and fixes

The file is rejected before conversion. ffprobe could not read it as audio. It may be damaged, a video with no audio track, or not really the format its extension claims. Open it in a player to confirm.

The saving is small or zero. The source is already at or below your target. Lower the bitrate, switch to mono, or accept that the file is already small. Re-encoding a 96 kbps file to 96 kbps saves nothing.

The output is larger. The target was above the source bitrate, or the format change added overhead. Read the reported reason, then choose a lower rate or keep the original format.

Speech sounds hollow at 64 kbps. Try 80 or 96 kbps mono. Bitrate needs vary with the recording: clean studio speech tolerates lower rates than a noisy room recording.

Stereo music sounds narrow in mono. Mono sums both channels, which can cancel parts of a wide mix. Use mono for speech, not for music you care about.

High sample rates did not shrink much. The bitrate does most of the work. Dropping 96 kHz to 48 kHz helps, but it cannot make up for a high bitrate target.

A long file times out. Encoding has a time limit. Split the audio first or convert in shorter segments. The 50 MB upload cap also limits very long uncompressed files.

Cover art disappeared. The output is audio only. Embedded artwork and video tracks are intentionally dropped.

The result will not play on an old device. Some players do not support Opus or certain AAC profiles. Choose MP3 for the widest compatibility.

#Choosing a format

• MP3 — universal support, predictable sizes, fine at 128–192 kbps for music.
• AAC (.m4a) — better quality than MP3 at the same bitrate, especially below 128 kbps. Native on Apple devices.
• Opus — the most efficient of the lossy options at low bitrates. Excellent for speech and streaming, supported by modern browsers and players.
• FLAC — lossless, roughly half the size of WAV, no quality loss. Good for archiving.
• WAV — uncompressed, large, best for editing rather than distribution.

If the destination is a podcast host, check its recommended format before converting. If it is a phone from a decade ago, MP3 is the safe answer.

#Quality checks

1. Listen to a quiet passage and a loud one. Artefacts show up in cymbals, sibilance and background noise before they are obvious in speech.
2. Compare the waveform or spectrum if it matters. A spectral view shows where the high frequencies were cut.
3. Check duration and channels. The result reports both. A duration change suggests a container or decode problem.
4. Do not re-encode repeatedly. Each lossy generation loses a little more. Keep the best source and encode once.
5. Keep the original. Storage is cheap; a ruined source is not recoverable.

#Reducing size across other formats

Audio is not the only place where the right target beats a blind "maximum compression" setting. The image compressor works the same way: choose a quality level that suits the destination instead of shrinking until the result looks bad. The PDF to PNG converter is the audio equivalent of choosing a resolution: 150 DPI is right for a screen, 300 for print, and picking 300 for a thumbnail just makes a bigger file.

#Privacy

The file is uploaded to the server and encoded there with ffmpeg. Temporary copies are deleted afterwards and download links expire. That is normal for server-side media conversion, but it is still an upload: do not send private voice notes or unreleased recordings unless you are comfortable with that.

#Frequently asked questions

#Bottom line

The useful question is not "what is the smallest file?" but "what is the smallest file that still sounds right for this use?" Pick the format for compatibility, the bitrate for the content, and mono for speech. Start with the audio bitrate reducer, and apply the same target-driven thinking to images with the image compressor.