Audio Trimmer

The Audio Trimmer cuts one section out of an audio file, using a start time and an end time you type in, and lets you download it. For example, you can take seconds 10 to 30 out of an hour-long recording. It uses a fast mode that does not re-encode the audio, so quality is not reduced, but that comes with a few things worth knowing up front: it extracts a single continuous range only, there is no waveform to drag, no fade in or out, no joining of several parts, and no format conversion. The output file has the same format as the file you uploaded.

How to use it

1. Click the dashed box labelled 'Click to upload audio' and choose an audio file. The box then shows the file name.
2. In 'Start time (format 00:00:00)' enter where the kept section begins, for example 00:00:10. In 'End time (format 00:00:00)' enter where it ends, for example 00:00:30. Both fields are required.
3. Click 'Start Trimming'. On the first use the button reads 'Loading processing core…', then changes to 'Trimming…' once the core is ready.
4. When it finishes, a player appears so you can listen to the result. If it sounds right, click 'Download Trimmed Audio'. The downloaded file name is the original name with trimmed_ in front.

A usage scenario

Say you have an interview recording called interview.mp3 and want to keep only the answer between 10 and 30 seconds. Upload it, type 00:00:10 as the start and 00:00:30 as the end, and click 'Start Trimming'. You get a file of roughly 20 seconds named trimmed_interview.mp3. Because nothing is re-encoded, the real cut points snap to nearby boundaries in the encoded data, so the actual start and end can differ from what you typed by tens to a few hundred milliseconds. The note at the top of the page says the same. If a word must not be clipped, widen the range a little. The exact error depends on how each file was encoded, so we do not quote a specific figure.

Supported input and known limits

• The file picker only offers audio files. Whether a file can actually be processed depends on whether the bundled ffmpeg core recognizes it; if it does not, you will see the failure message.
• Times must follow the 00:00:00 (hours:minutes:seconds) pattern shown on the page. If the format is wrong, the end is not later than the start, or the range falls outside the audio, a red message starting 'Processing failed. Please make sure the time format is correct…' appears.
• If either time field is empty, clicking 'Start Trimming' shows the message 'Please enter both a start time and an end time.' and does not start processing, so fill in both first.
• The output has the same format as the input, because the result is saved with the original file's extension. To change format, use the site's Audio Format Converter.
• Not re-encoding means cuts can only land on boundaries in the encoded data, so precision is limited. Leave a margin when accuracy matters.
• The whole file is read into browser memory. There is no size limit in the code, but a very large file can slow the browser or fail.
• You can extract one range per run. To keep several parts, trim them one at a time.
• The first use needs a network connection to download a core of roughly 25-30MB, so it will not work offline or behind a restrictive network.

FAQ: Will the clip sound worse than the original?

No. The tool copies the data directly (ffmpeg's -c copy mode) without decoding and re-encoding, so the kept part has exactly the same quality as the source. The trade-off is limited cut precision.

FAQ: Why is the first run slow?

The browser first has to download the ffmpeg processing core, about 25-30MB, and how long that takes depends on your connection. Once loaded it is reused while the page stays open, so later trims do not download it again.

FAQ: What should I do when it says processing failed?

Check that the times look like 00:00:10, that the end is really later than the start, and that neither goes past the length of the audio. If those are fine, try another file, or convert the audio to MP3 with the format converter and trim that.

FAQ: Can I use it on video files?

The upload box is meant for audio files and the tool is not designed for video. For video work, use the other video tools on the site.

Privacy and processing

Trimming happens in your browser. The audio is written to a virtual file system inside the page and processed by a WebAssembly build of ffmpeg, and the code has no step that uploads your audio to a server. The only network traffic is the first-time download of the ffmpeg core from unpkg.com, a public CDN, and that request does not contain your file. The result is played and downloaded through a temporary local link that is released when you switch files or leave the page.