Audio trimmer

What this tool does

It opens an audio file (MP3, WAV, M4A, OGG, Opus, FLAC or WebM, as far as your browser can decode it), draws its waveform, and lets you keep only the part you want. You set the start and the end by dragging two handles, with the arrow keys, or by typing times to the millisecond. You can add a fade in and a fade out, listen to the result, and save it as a WAV file. Everything happens in your browser with the Web Audio API: the file is not uploaded anywhere.

How to use it

1. Open a file with the picker or drop it on the page. The tool shows its length, sample rate and number of channels.
2. Choose the part to keep. Drag the [ and ] handles on the waveform, or type the start and end in seconds (12.5) or minutes (1:02.250). On a focused handle, the arrow keys move it by 0.1 second, or by 1 second with Shift.
3. Add fades if you need them. A fade of 0.01 to 0.05 seconds removes the click you can hear when a cut lands in the middle of a sound; a fade of a few seconds makes a song end gently. Equal power sounds smoother than linear for long fades.
4. Listen with Play selection: you hear exactly what will be saved, fades included.
5. Download WAV. The tool shows the file size before you save. 16-bit is CD quality and enough for almost everything; 24-bit keeps more detail for further editing.

Why WAV, and why the sample rate is kept

Saving as WAV means the samples you hear are written to the file as they are: no second round of lossy compression, so nothing gets worse. MP3 and AAC encoders also add a little silence at the start and pad the end to a whole frame, which is why a clip cut and re-encoded as MP3 can start late or leave a gap when looped. WAV has no frames and no padding: the cut is exactly as long as you chose. The longer explanation, with numbers, is in our article on trimming audio without re-encoding artefacts.

Browsers decode audio at the rate of the audio context and resample anything else. The tool reads the sample rate from the file's header (WAV, MP3, FLAC, Ogg) and decodes at that rate, so a 44.1 kHz file is saved at 44.1 kHz and the samples are not resampled. For formats it cannot read the rate from, such as M4A, it decodes at 48 kHz. Above 96 kHz, which not every browser can handle, the file is decoded at 96 kHz and the tool says so.

Size and memory

WAV size is simple arithmetic: sample rate × channels × bytes per sample × seconds, plus a 44-byte header. Thirty seconds of 44.1 kHz stereo at 16 bits is 44,100 × 2 × 2 × 30 + 44 = 5,292,044 bytes, about 5.3 MB. That is roughly five times an MP3 at 192 kbit/s, so for sharing a long recording you may want to compress the WAV afterwards in an app you trust. While editing, the decoded audio is held in memory as 32-bit floating point numbers (4 bytes per sample per channel), so the tool accepts recordings up to about 26 minutes of 48 kHz stereo; longer ones should be split first.

Tips

• Zoom with your ears, not your eyes: play the selection and nudge the handles by 0.1 s until the cut sounds clean.
• Cut in a quiet spot, between words or beats, and a 10 ms fade is usually enough.
• To make a ringtone, keep 20 to 30 seconds and add a one-second fade out.
• Need the recording first? The voice recorder records your microphone in the browser; the file hash checker can confirm two exports are byte-identical.