Changing audio file size without losing quality means multiplying each decoded sample by a chosen gain and writing the result to a lossless 16-bit PCM WAV container, which is exactly what the Volume Changer tool does. Volume Changer runs entirely in the current browser tab, so the source file is never uploaded and never re-encoded through a lossy codec. The percentage you select is a direct linear amplitude multiplier: at 100% samples keep their decoded value, at 200% each sample is doubled, and at 0% the file becomes silence while its duration, sample rate, and channel count stay identical to the source. Because the exported file uses the standard RIFF WAVE layout with format code 1 and 16-bit signed PCM, the output is broadly inspectable, preserves the original sample rate and channel layout, and matches the asymmetric range where negative full scale maps to minus 32768 and positive full scale maps to 32767. The tool also reports a peak value after gain and a clipped-sample count, which makes it possible to see whether the chosen gain pushed any samples past digital full scale and whether audible distortion is likely. That is the practical answer for anyone who needs to raise or reduce the level of an audio file and still ship a clean, untouched-by-recompression WAV.
What "changing file size without losing quality" means for audio
For audio, the phrase "change file size" usually means one of three things: resampling to a different bit rate, recompressing through a smaller lossy codec, or changing the loudness level of the waveform. The first two almost always lose quality because they throw away samples or frequency information, and a second generation of MP3 or AAC applied to an already-encoded file compounds that loss. The third — adjusting the loudness of the file itself — does not require any recompression if the operation is done by multiplying decoded samples and writing them back to a lossless container. That distinction is the foundation of the Volume Changer workflow, and it is also why the tool always exports WAV rather than MP3 or AAC.
Volume Changer uses the browser's Web Audio decoder to convert the complete encoded file into floating-point channel samples in memory, as described in the W3C Web Audio API specification for decodeAudioData. The decoder returns a complete PCM buffer for the full duration of the file, so no portion of the audio is approximated, time-stretched, or resampled for the gain step. The gain is then applied sample by sample to that buffer, and the result is encoded as a fresh RIFF WAVE file using the structure documented in the Microsoft Learn reference for WAVEFORMATEX. Because the same uncompressed path is used on the way in and on the way out, no generation of lossy compression is introduced.
How Volume Changer preserves audio quality during a volume edit
The tool preserves quality in three concrete ways. First, decoding happens locally, so the original bit stream is never uploaded and the file never passes through a server that could re-encode it. Second, the gain itself is a linear amplitude multiplier, not a dynamic compressor, limiter, or loudness normalizer, which means the relative shape of the waveform is preserved — only the scale changes. Third, the output uses the standard RIFF container with format code 1, the original sample rate, the original channel count, and 16-bit signed PCM, channel samples interleaved frame by frame.
Samples that, after multiplication, exceed the digital full-scale range of minus 1 or plus 1 are clamped to that range, and the tool counts how many finite samples had to be clipped. A nonzero clipped-sample count is a sign that the chosen gain may have introduced audible distortion, although zero clipping does not by itself guarantee ideal perceived loudness, because perceived loudness also depends on frequency balance, duration, dynamics, speakers, and listening conditions. For a related practical scenario where the goal is editing a file rather than relying on playback controls, the How to Change Volume on AirPods by Editing the File walkthrough covers the same approach.
Adjust audio volume without losing quality: the steps
- Open Volume Changer and prepare a browser-decodable audio file no larger than 50 MiB.
- Select the file from your device and wait for the file name to appear.
- Choose a linear output volume between 0% and 400%; 100% keeps the original amplitude, 200% doubles it, and 0% writes silence while preserving duration and channels.
- Click Change volume and wait for the processing to finish — decoding, gain application, clipping detection, and WAV encoding happen in the current tab.
- Preview the resulting WAV through suitable headphones or speakers and listen to the complete output, not just a fragment.
- Inspect the reported peak value and the clipped-sample count before downloading; if clipping is nonzero and distortion is audible, lower the percentage and reprocess.
- Download the file, which will be saved with a filename ending in -volume-adjusted.wav, and verify its duration and channel count before replacing any production audio.
Volume percentage and what it does to your samples
The percentage is a direct linear amplitude multiplier applied to every finite decoded sample. At 100%, samples keep their decoded amplitude exactly. At 50%, each sample is multiplied by 0.5, which approximately corresponds to a 6.02 dB reduction in amplitude. At 200%, each sample is multiplied by 2, which is approximately a 6.02 dB amplitude increase before clipping. At 400%, each sample is multiplied by 4, which is approximately a 12.04 dB amplitude increase, and many signals will be heavily clipped. These figures come from the standard relation that doubling linear amplitude is a 6.02 dB change, and the same math is why the tool reports a peak and clipped-sample count rather than a dB slider.
For a concrete worked example, take a decoded sample whose amplitude sits at 0.5 in the floating-point range. Multiplied by a 200% gain, it becomes 1.0, which is exactly at positive digital full scale and not clipped. A sample originally at 0.6 multiplied by 200% becomes 1.2, which is clamped to 1.0 and counted as one clipped sample in the report. That single arithmetic step — original value times gain, clamped to plus or minus 1.0 — is the whole operation.
| Selected percentage | Multiplier applied to each sample | Approximate effect |
|---|---|---|
| 0% | × 0 | Full-duration silence, original sample rate and channel count preserved |
| 50% | × 0.5 | Quiet version of the source, near 6 dB lower amplitude |
| 100% | × 1 | Source amplitude unchanged |
| 200% | × 2 | Louder version, near 6 dB higher amplitude; clipping possible on peaks |
| 400% | × 4 | Heavily boosted version, near 12 dB higher amplitude; clipping likely |
File, duration, and channel limits to know before uploading
The tool enforces several interacting limits that exist to bound memory allocation and main-thread work on ordinary laptops and mobile devices. The selected file must be at most 50 MiB. The decoded duration must be at most five minutes. The channel count must be at most eight, which covers mono through 7.1 surround. The total number of decoded channel samples must be at most 30 million. Because a compressed file can expand substantially after decoding, the decoded limits matter even when the selected file is small — for example, a short MP3 with a low bit rate can decode to a very large PCM buffer relative to its encoded size. The tool therefore checks the decoded PCM buffer, not just the on-disk file size.
Decoding depends on the browser, operating system, and codec build, and a filename extension is not proof that the file can be decoded. Supported families include MP3, WAV, M4A, AAC, Ogg, WebM, and FLAC, as long as the browser can decode them. If decoding fails, the tool stops and shows an error rather than producing an empty or partial output. The exported file is always WAV regardless of source format, which is a deliberate choice: PCM16 WAV can be encoded locally without adding a new lossy codec dependency and exposes a deterministic complete result that any audio editor can inspect. The downside is that WAV is usually much larger on disk than MP3, AAC, or Ogg, so the output trades file size for transparency.
Practical tips for clean, undistorted volume edits
Start near 100% whenever you are unsure, and adjust in small steps if you need more loudness. Always preview the full output rather than a short fragment, because distortion often shows up only on the loudest peaks of the source. Read the clipped-sample count after each change; if it is nonzero and you can hear distortion, reduce the percentage and reprocess. Remember that the operation cannot restore information that was already clipped in the source, and repeated PCM conversions can complicate version control, so keep the original file alongside the edited one. The tool does not preserve source tags, album art, chapters, loop points, cue markers, encoder metadata, or the original compression settings, so do not rely on the output to carry any of that metadata. If your goal is professional delivery, use a measured audio workflow with true-peak metering and platform-specific loudness targets — this tool is a linear gain stage, not a mastering processor, and it does not calculate LUFS, true peak, replay gain, or broadcast compliance. For most everyday tasks such as boosting a quiet voice memo, lowering an overcooked recording, or silencing a section, the linear gain path plus the PCM16 WAV export is enough to keep quality intact.