Before using Audio Effects Online, confirm that your file is browser-decodable and no larger than 50 MiB, and note that both decoded input and complete output are limited to 30 million channel samples.

Those checks matter because a failed decode or an output-budget failure produces no downloadable result. Privacy is another important consideration: file reading, decoding, processing, WAV encoding, and download creation all happen in the current browser tab. Audio is not uploaded to Lizely or stored in an account, and there is no server render or hidden AI transformation.

what should i know before i use audio effects online
Before You Use Audio Effects Online: A Pre-Use Checklist

Check compatibility and processing limits before you start

Compatibility is the first gate, and a recognized filename does not guarantee that a file will decode. The tool accepts names associated with MP3, WAV, M4A, AAC, Ogg, WebM, and FLAC, but actual codec and profile support varies by browser and operating system. Input decoding uses the browser's Web Audio implementation, whose decoding contract is defined in the W3C Web Audio specification.

Check Required value Why it matters
Encoded source One browser-decodable audio file, 50 MiB or less The size limit applies to the encoded file before it is decoded.
Filename types MP3, WAV, M4A, AAC, Ogg, WebM, or FLAC names are accepted Acceptance of a filename does not guarantee codec support in the current browser.
Channel structure 1–8 equal-length channels Audio outside this channel range cannot pass the local pipeline.
Sample rate 8,000–192,000 Hz The decoded sample rate must remain within the supported range.
Decoded input No more than 30 million channel samples This budget limits the total decoded data, not just the encoded file size.
Complete output No more than 30 million channel samples Any effect tail is included, so the output can exceed the input budget.
Effect amount A whole number from 0 through 100 The selected percentage controls the blend or scale of the chosen effect.

A channel sample is one sample value at one point in one channel. The total is the number of sample frames multiplied by the number of channels. For echo and reverb, the complete delayed tail is added to that calculation. If the finished output would exceed 30 million channel samples, nothing is truncated and no WAV is created.

Know exactly what each effect will change

The amount is specific to the selected operation rather than a universal “intensity” control. Each option has a disclosed, deterministic behavior, and only one effect is applied per run.

Effect Exact behavior Important boundary
Echo At 100%, every input sample is retained and a half-strength copy is added 200 milliseconds later. Lower amounts scale the delayed copy linearly. The full delayed tail is retained, but overlapping dry and delayed samples can clip.
Lightweight reverb Copies are added at 29.7, 37.1, and 41.1 milliseconds with weights of 0.50, 0.35, and 0.25 multiplied by the selected amount. This is an early-reflection design, not a room or professional processor emulation.
Reverse The decoded sample frames are read in the opposite order and blended with the original according to the amount. It reverses samples rather than playing the file backward as a new recording.
Fade in A linear fade scales from the selected dry proportion at the beginning to full level at the end. The amount controls the dry proportion at the starting edge.
Fade out The inverse linear fade moves from full level to the selected dry proportion. The amount controls the dry proportion at the ending edge.
Normalization The largest absolute decoded sample is used to scale the result toward a 0.95 sample peak. It does not measure perceived loudness, LUFS, dynamics, or inter-sample true peak.

The reverb coefficients are fixed product choices, not measurements of a concert hall, plate, spring, studio, or branded device. The effect does not model room dimensions, frequency-dependent decay, stereo diffusion, damping, pre-delay, convolution, or a professional reverberation tail. It can add space or repetition for a quick sound-design draft, but it should not be presented as an accurate acoustic simulation.

Normalization has a similarly narrow purpose. It adjusts the sample peak, not the perceived volume of different songs or voices. It does not provide equal loudness across recordings, broadcast compliance, or dynamics matching. Silent input remains silent.

Use the browser workflow from source to download

  1. Keep an untouched copy of your source, then open Audio Effects Online and choose one browser-decodable audio file no larger than 50 MiB.
  2. Select echo, lightweight reverb, reverse, fade in, fade out, or normalization. Set a whole-number amount from 0% through 100%.
  3. Apply the selected effect locally. The tool calculates the complete output before encoding it; if decoding fails or the complete result exceeds the output budget, no result file is created.
  4. Inspect the reported channel count, sample rate, input frame count, output frame count, output peak, and exact WAV byte size. Confirm that the displayed result corresponds to the current file, effect, and amount.
  5. Download the new PCM16 WAV, compare it with the untouched source, and retain both files so you can revert or try a different setting without repeating earlier work.

Read the output numbers in context

Compare the input and output frame counts before deciding whether the result is complete. Echo and reverb can make the output longer because their full tails are included. The tool does not cut off the last echo to force a file through the limit; it either produces the complete calculated result or produces none.

Review the output peak as well. Strong echo or reverb settings can push overlapping samples beyond the legal range. Finite output samples are constrained to the -1 through +1 floating-point range before WAV quantization, so checking the peak and listening for distortion are both useful. A normalized result targets a 0.95 sample peak, but that figure does not guarantee that true peaks between samples remain below a digital clipping boundary.

Changing the file, effect, or amount invalidates older work. Use the numbers attached to the current result rather than a preview or download produced from earlier selections. For a final check, listen for unexpected clipping, abrupt fade edges, unwanted repetition, or a reversal that does not match your intended edit.

Plan for privacy, format changes, and chained effects

The download is a new interleaved, little-endian PCM16 RIFF/WAVE file. It preserves the processed channel samples and sample rate, while floating-point samples are quantized to signed 16-bit PCM. Microsoft's WAVEFORMATEX documentation defines PCM block alignment as channels multiplied by bits per sample divided by eight; at 16 bits, that is two bytes per mono frame and four bytes per stereo frame. The page reports the exact byte size after processing.

Do not expect the export to preserve the source codec, bitrate, compression quality, tags, artwork, chapters, cue points, loop markers, or other container metadata. If those parts of the original file matter, keep the source separately. The untouched original also provides the clean reference for judging the processed WAV.

Building a chain requires separate runs. Download the first result, inspect it, and then use that WAV as the input for a later effect. Repeated lossy decoding and PCM processing can reveal source artifacts, while repeated clipping or normalization can alter dynamics further. Keeping the original makes each stage easier to compare and prevents a processed file from becoming your only source.