Why is my file bigger after reducing it?

The output is always an uncompressed 16-bit PCM WAV, so if your source was already a compressed MP3, M4A, or OGG, the WAV can come out larger even at a lower sample rate. This tool saves the most when your input is a big WAV, or when you specifically need WAV output. The before-and-after size shown on screen lets you confirm you actually saved space before keeping the file.

What sample rate should I use for a voice recording?

16000 Hz is the sweet spot for speech: it preserves everything up to 8 kHz, well above the ~3.4 kHz that carries voice intelligibility. Drop to 8000 Hz for the smallest, telephone-grade result. Avoid these low rates for music, which needs 22050 Hz or higher to keep cymbals and high detail from sounding dull.

How much does converting to mono actually save?

Mono collapses the left and right channels into one, which halves the audio data compared to stereo. For voice memos and most single-source sound effects there's no meaningful stereo image, so you lose nothing you'd hear. It's on by default for that reason.

Can I predict the output size before running it?

Yes, because a WAV's size is fixed math, not dependent on the audio's complexity: sample rate x channels x 2 bytes x seconds. A one-minute mono clip at 16000 Hz is about 1.9 MB, and halving the rate to 8000 Hz roughly halves that. This makes it easy to target a size.

What audio formats can I feed it?

Anything your browser can decode and play, which typically includes MP3, M4A/AAC, OGG, FLAC, and WAV. Decoding happens on your device using the browser's audio engine, and nothing is uploaded. The result always comes back as a WAV at your chosen sample rate.

Audio Size Reducer

Compress audio file size.

Choose an audio file

Convert to monoSample rate (Hz)

What is Audio Size Reducer?

Audio Size Reducer is a free tool that compresses an audio file's size by lowering the sample rate and mixing down to mono, right in your browser. It shows the original and reduced sizes side by side — useful for voice notes, sharing, and game sound effects.

Key features

Mono conversion halves the data versus stereo
Selectable sample rates — 16000 Hz suits speech well
Shows original and reduced file sizes for comparison
Outputs a WAV at your chosen sample rate

How to reduce audio file size

This tool makes audio smaller by throwing away data your ears don't need for the job at hand — extra channels and high frequencies. It's built for voice memos, spoken-word clips, and game sound effects, where a smaller file matters more than studio fidelity. Everything runs in your browser: the file is decoded, downsampled, and re-encoded on your device, and nothing is uploaded.

The two levers you control

There are exactly two settings, and each one cuts the size in a predictable way.

Convert to mono (on by default) collapses left and right channels into one. A stereo file carries two independent tracks, so mixing to mono halves the audio data outright. Voice notes and most single-source sound effects have no meaningful stereo image, so you lose nothing audible.
Sample rate decides how many times per second the waveform is measured. The choices are 8000, 11025, 16000, 22050 (the default), and 32000 Hz. Lower is smaller.

Picking a sample rate

A sample rate can only reproduce frequencies up to half its value (the Nyquist limit), so 16000 Hz preserves everything up to 8 kHz — well above the roughly 3.4 kHz that carries speech intelligibility, which is why 16000 Hz is the sweet spot for narration and voice memos. Drop to 8000 Hz for telephone-grade voice at the smallest size; music and anything with cymbals or sparkle will sound dull because those live above 8 kHz. Stay at 22050 or 32000 Hz when you want the clip to still sound reasonably full.

How the math works

The output is an uncompressed 16-bit PCM WAV, and its size is completely predictable: sample rate × channels × 2 bytes × seconds. A one-minute clip at 16000 Hz mono is 16000 × 1 × 2 × 60 ≈ 1.9 MB; halving the rate to 8000 Hz halves that to about 960 KB. Because the size is fixed by these numbers rather than by the audio's complexity, you can estimate the result before you even run it.

The gotcha worth knowing

The output is always a WAV, which is uncompressed. If your source is already a compressed MP3, M4A, or OGG, converting it to WAV can produce a larger file unless you cut the sample rate and mono aggressively — a 128 kbps MP3 is far denser than PCM audio. This tool shines when you need WAV output specifically, when you're shrinking an already-large WAV, or when telephone-quality voice at a tiny size is the goal. The tool decodes any format your browser can play, and it shows the original and result sizes side by side so you can confirm you actually saved space before keeping the file.

Frequently asked questions

Why is my file bigger after reducing it?: The output is always an uncompressed 16-bit PCM WAV, so if your source was already a compressed MP3, M4A, or OGG, the WAV can come out larger even at a lower sample rate. This tool saves the most when your input is a big WAV, or when you specifically need WAV output. The before-and-after size shown on screen lets you confirm you actually saved space before keeping the file.
What sample rate should I use for a voice recording?: 16000 Hz is the sweet spot for speech: it preserves everything up to 8 kHz, well above the ~3.4 kHz that carries voice intelligibility. Drop to 8000 Hz for the smallest, telephone-grade result. Avoid these low rates for music, which needs 22050 Hz or higher to keep cymbals and high detail from sounding dull.
How much does converting to mono actually save?: Mono collapses the left and right channels into one, which halves the audio data compared to stereo. For voice memos and most single-source sound effects there's no meaningful stereo image, so you lose nothing you'd hear. It's on by default for that reason.
Can I predict the output size before running it?: Yes, because a WAV's size is fixed math, not dependent on the audio's complexity: sample rate x channels x 2 bytes x seconds. A one-minute mono clip at 16000 Hz is about 1.9 MB, and halving the rate to 8000 Hz roughly halves that. This makes it easy to target a size.
What audio formats can I feed it?: Anything your browser can decode and play, which typically includes MP3, M4A/AAC, OGG, FLAC, and WAV. Decoding happens on your device using the browser's audio engine, and nothing is uploaded. The result always comes back as a WAV at your chosen sample rate.

Privacy

Resampling and downmixing to mono are done locally; the original audio file never leaves your browser.