Cutting the audio out of a video file and saving it as a WAV happens entirely inside your browser when you load a browser-decodable source into Audio Cutter, enter a start time and a later end time, and export the resulting range. The tool uses the Web Audio API's decodeAudioData method to turn the audio portion of your file into an in-memory AudioBuffer, rounds your two time inputs to the nearest decoded sample frame, and writes a brand-new uncompressed PCM16 WAV file that you download directly from the page. Nothing is uploaded to a conversion server, and the source file is never copied off your device. Whether the file you start with is a video container that carries an audio track (such as a WebM) or an extracted audio track (MP3, WAV, M4A, AAC, Ogg, or FLAC), the result is the same: a frame-accurate slice of the original waveform saved as 16-bit signed little-endian PCM samples with a standard 44-byte RIFF/WAVE header.
Why Use a Browser Tool to Cut Audio from Video
Most desktop video editors can extract and trim audio, but they require installation, account logins, or codec packs. A browser-based audio cutter flips that around: the page loads, you pick the file, and the entire decode-cut-encode pipeline runs in the current tab. Audio Cutter uses the Web Audio API to decode the source, so the moment the buffer is ready it displays the decoded duration, sample rate, channel count, and total frame count. That four-number readout lets you verify the audio actually loaded before you decide where to cut. If the value seems off, you can pick a different file immediately without losing progress, because choosing another source clears the old buffer, the source player, and any prior result.
Because the work is local, you keep control of your source. No compressed source is transmitted to a server, no decoder is silently substituted, and the exported file is the only artifact your machine writes. For anyone who needs a clean WAV from a specific portion of a video's audio — a single interview answer, one musical phrase, a voiceover line, or a sound effect — this workflow replaces the "open a video editor, render to WAV, hope the format matches" routine with a predictable three-field form: pick, start, end.
What "Cut" Means Here: Frame-Accurate Range Selection
A "cut" in Audio Cutter is not a rough timer estimate. When you enter a start time and an end time, those values are converted to the nearest decoded sample-frame indices using the AudioBuffer's sample rate. The Web Audio API may resample your source to the AudioContext sample rate during decode, so the displayed sample rate describes the working buffer, not necessarily the original encoded rate. The start frame is included in the output; the end frame is excluded. That exclusive boundary means a 0.000-second selection rounded to less than one full frame is rejected instead of producing an empty file.
Start must be a non-negative number, end must be later than start, and neither value can exceed the exact decoded duration. Because production code and isolated tests share the same frame-range function, the duration shown above the download button is the selected frame count divided by the decoded sample rate — not a separate timer estimate. The displayed values will agree with the WAV bytes you receive, which removes the usual guesswork when trimming a clip for sync work.
How to Cut the Audio from Your Video File
Two paths lead to a usable cut, depending on what kind of file you start with. If your video is a WebM, the file picker may recognize it directly; Web Audio may decode the audio track from a WebM container, but recognition does not guarantee that every codec variation can be decoded, and the video frames are not part of the result. For MP4, MOV, MKV, and most other containers, extract the audio track first using your operating system or an offline tool so you end up with an MP3, M4A, WAV, Ogg, or FLAC file that the picker is designed to recognize; actual decoding still depends on your browser. Once you have an audio file in hand, the cutting workflow is identical.
- Open Audio Cutter and click the file picker. Choose a browser-decodable audio file (or a WebM you want to treat as audio) that stays within the 25 MB encoded size limit. The picker checks the file name and MIME type before reading.
- Wait for Web Audio to finish decoding. The page will then display the decoded duration, sample rate, channel count, and frame count. If you see a decode error instead, your file is unsupported, damaged, mislabeled, encrypted, or incomplete — pick a different source.
- Enter a non-negative start time and an end time that is later than start and no greater than the decoded duration. Both fields are converted to the nearest sample frame before cutting.
- Select Cut and export WAV. The tool clamps each Float32 sample in the selected range to the -1 through 1 PCM range, writes negative full scale as -32768, positive full scale as 32767, and treats any non-finite values as silence.
- Preview the resulting PCM16 audio in the page's result player. Listen for the spoken words, transient, or breath at both boundaries to confirm the cut lands where you intended.
- Download the WAV. The reported output duration and file size match the bytes about to be saved, so you can sanity-check the result before writing it to disk.
If you plan to put the trimmed audio back onto a video later, a separate local workflow for merging clips and tracks is documented in Add Audio for Video Free: Merge Clips Locally, which covers the matching track-merge step without re-uploading either source.
What the Downloaded WAV Contains
Every download is a freshly encoded RIFF/WAVE file. The 44-byte header records PCM format 1, the channel count, the decoded sample rate, the byte rate, the block alignment, 16 bits per sample, and the exact data size. The audio body is interleaved, little-endian, signed 16-bit PCM in the original channel order. Multiple channels are kept separate and interleaved in their decoded order — stereo stays stereo, and sources with up to eight channels are preserved.
What the WAV does not carry is just as important to know. It does not retain the source's MP3, AAC, Vorbis, Opus, FLAC, or other compressed encoding. It does not preserve ID3 tags, Vorbis comments, album art, cue points, chapters, loudness metadata, encoder settings, or other container fields. The result is a clean, portable audio file that any media player or editor can open.
| Source property | Preserved in the WAV export? |
|---|---|
| Audio samples in the selected range | Yes, after clamp to the PCM range |
| Channel count and channel order | Yes, up to the eight-channel limit |
| Sample rate of the decoded buffer | Yes, written into the WAV header |
| Original compressed codec and bitrate | No — WAV is uncompressed PCM16 |
| Tags, album art, chapters, cue points | No — not part of the WAV format |
| Loudness or encoder metadata | No |
Because PCM WAV is uncompressed, the file can be much larger than a short MP3 or AAC slice of the same audio. Its data size equals the selected frame count multiplied by the channel count and by two bytes per sample, plus the 44-byte header. For a worked example, a 10-second stereo cut at a 44100 Hz decoded rate contains 10 × 44100 = 441000 frames per channel, so the data is 441000 × 2 channels × 2 bytes = 1,764,000 bytes of PCM, and the complete WAV on disk is 1,764,000 + 44 = 1,764,044 bytes, or roughly 1.68 MB.
Limits, Decode Errors, and When to Use a Dedicated Editor
The browser enforces hard limits rather than silently degrading. The encoded input file must be 25 MB or smaller. After decoding, the audio must be 15 minutes or shorter, use no more than eight channels, no more than 192 kHz, and no more than 30,000,000 channel-samples in total. The decoded-sample cap means a high-rate or multichannel file can hit the boundary well before the duration limit, and the tool reports this with a clear rejection message rather than producing a partial file. Browser decoding has to happen before the decoded duration can be inspected, so a heavily compressed long file still consumes temporary memory during the decode attempt.
| Limit | Threshold | When it applies |
|---|---|---|
| Encoded file size | 25 MB | Before decode |
| Decoded duration | 15 minutes | After decode |
| Channel count | 8 channels | After decode |
| Sample rate | 192 kHz | After decode |
| Total channel-samples | 30,000,000 | After decode |
Several capabilities are deliberately out of scope. The tool does not offer a target resampling control, a remix, normalization, fades, noise removal, pitch shifting, or compression of the WAV. It does not install codecs and does not send your file to a conversion server. Audio contexts are closed after decoding and during cleanup, and in-flight reads and decodes are guarded so an older job cannot overwrite a newer selection. Changing either time field clears the previous WAV, and cutting again replaces and releases the prior download URL.
For dependable work, preview the cut WAV, confirm the spoken words or musical transient at both boundaries, and keep the original file. If you need waveform-level fades, lossless compressed-frame cutting, metadata preservation, or production mastering, switch to a dedicated editor that targets those specific outcomes — the value here is a fast, local, frame-accurate WAV from one continuous range, not a full audio production suite.
For a deeper look, see Add Reverb to a Video's Extracted Audio Track.