Audio Trimmer in the browser is not a generate
/free-tools/audio-trimmer never spends a credit. Do not pay a model to do an in-browser file job.
/free-tools/audio-trimmer never spends a credit. Do not pay a model to do an in-browser file job.
Audio Trimmer loads a file, draws a waveform from the decoded samples, lets you set in and out, previews the slice, and exports 16-bit PCM WAV of just that region. It runs in your browser. Nothing is uploaded. Nothing is sent to a model. There is no account gate and no generation meter. That is the product constraint for the whole free-tools shelf: the File API and Web Audio on the visitor's machine. A "trimmer" that quietly called a video model, a voice row, or a chat completion would be a generate wearing a utility label.
Versely has no free tier for generation. Every model call costs credits. Browser utilities sidestep that because they cost nothing to run. Use this tool when the job is "cut this take to the line." Do not prompt a video or audio model to "shorten this file."
In and out are a waveform, not a prompt
The file you choose is decoded into memory locally, sliced there, and offered back as a download. Network off, it still works. That is also why it is actually free: no hardware of ours is in the loop. What it costs is the other book — per-second video, per-character voice, per-megapixel stills, caption burns, editor exports. Audio Trimmer is not in that book.
The waveform is drawn from the decoded samples, not from a fake placeholder. In and out are range sliders. Play selection previews the slice. Download writes a WAV named from the source plus -trim. Do not paste the take into an AI video generator and ask for a "shorter version." You will spend a clip rate to invent a new performance. You will not get a sample-accurate cut.
The guard is 80 MB and 20 minutes once decoded. Compressed files expand into PCM in memory, so a long podcast is heavy even when the mp3 looks small. That is a browser-memory cap, not a credit cap.
The export is WAV on purpose
The trim is not lossless. The browser has already decoded the source, and the export is 16-bit PCM WAV. You are not copying the original compressed packets, so this is a re-encode, not a lossless cut of the bitstream. That is also why a trimmed MP3 becomes a larger WAV. Browsers decode many codecs but ship no MP3/AAC encoder; uncompressed WAV is the format the tab can emit reliably. decodeAudioData also resamples to the AudioContext rate, often 48 kHz, so the exported WAV is not guaranteed to match the source sample rate.
The display downsamples to one min/max pair per pixel across the whole file, so it is an overview, not a zoomed editor. Quiet speech can look like a line even when the samples are there. The export still contains every sample in the selection.
If you needed loudness numbers on the same file, that is a different free tool on the same shelf — Loudness Meter — also in-browser, also zero credits, an estimate rather than a certified EBU R128 meter. If you needed to cut picture, Video Trimmer is the sibling file job. Neither is a generate.
Credits start when you invent a second clip
Caption burns, voice rows, editor exports: those quote a credit figure before they charge. Audio Trimmer does not quote a figure because there is no charge. Mixing the two is how teams burn a model on a file job — "just generate a tighter take." The tighter take is the in/out pair.
Already have the file and only need a shorter region → trimmer. Need a different performance → a generate, on purpose, billed.
FAQ
Does Audio Trimmer use credits or a free-generation quota?
Neither. It never spends a credit. The work never leaves your browser. Generation on Versely is a different surface and always bills.
Is the trim lossless?
No. The browser has already decoded the source, and the export is 16-bit PCM WAV. You are not copying the original compressed packets. A trimmed MP3 becomes a larger WAV for that reason.
Why does the waveform look flat on quiet speech?
The display downsamples to one min/max pair per pixel across the whole file. Quiet speech can look like a line even when the samples are there. The export still contains every sample in the selection.
Can I prompt a video model to cut the silence instead?
You can waste the credits. You will not get this tool's in/out pair. Animate a plate when the job is a plate. Trim a file here.