Audio Converter

Why the direction of a conversion matters

A WAV keeps every sample exactly as it was recorded. An MP3 encoder instead studies the signal, decides which parts a listener is unlikely to notice — quiet content masked by something louder, frequencies near the edge of hearing — discards them, and compresses the rest. That's the trick behind shrinking a 40 MB file to 4 MB, and it only goes one way.

This means the direction you convert in is not a minor detail. Going from lossless to lossy is normal and something you do once, at the end. Going lossy to lossless is harmless but recovers nothing you lost. Going lossy to lossy — an MP3 re-encoded into another MP3 — is the case to avoid, because the second encoder is making decisions about damage the first encoder already did, and the artefacts stack up audibly.

The practical rule: decode into a lossless format as soon as possible, keep everything lossless while you work, and only touch a lossy encoder once, right at the end, at the best bitrate your target allows.

What sample rate, bit depth and channels actually control

Sample rate is how many times per second the waveform gets measured, and it caps the highest frequency the file can represent — half the sample rate, known as the Nyquist limit. At 44.1 kHz that cap is 22.05 kHz, well past what people can hear, which is why the decades-old CD standard is still in daily use. Working at a higher rate during production has the same benefit as extra bit depth: more headroom while editing, not more detectable fidelity in the final file.

Bit depth sets how finely each sample is measured, which sets the noise floor. 16-bit yields roughly 96 dB of range, more than any home listening setup will use. 24-bit gives around 144 dB, useful mainly because it lets you record conservatively and still keep quiet passages well above the noise.

Channel layout is worth changing on purpose, not by default. Folding stereo into mono halves file size and guarantees the same result on single-speaker hardware, but it also adds the two channels together, so anything out of phase between them partially cancels. Listen to a mono version before you commit to shipping it.

• Leave the sample rate as-is unless a specific platform asks for something else
• Ship 16-bit; save 24-bit for masters and in-progress mixes
• 320 kbps MP3 for release copies, 192 kbps for previews, 128 kbps for speech
• Always listen to a stereo-to-mono conversion before you publish it

Why this runs entirely on your device

Most conversion sites upload your file to a server, process it remotely and hand back a download link. That means audio you may not have released yet — or a client's file, or something private — sits on infrastructure you have no control over, under a retention policy you've probably never read.

Nothing here gets uploaded. Your browser decodes the file with its built-in audio engine, resampling and encoding run in JavaScript on your own machine, and the result downloads straight from memory. It keeps working offline once the page has loaded, and there's nothing left on a server to worry about, because nothing was ever sent to one.

You might also need

• Volume NormalizerBring a track to a chosen LUFS target with a safe peak ceiling, then export a WAV.
• Audio Cutter & TrimmerSelect a region of a track, keep or delete it, and smooth the edges with fades.
• Audio AnalyzerGet a full technical read-out of a track: loudness, key, tempo, stereo width and spectral balance.