Audio Effects Online is a browser app to add reverb to audio that decodes one file in your tab, applies a disclosed three-tap reverb with a chosen amount from 0% to 100%, and downloads a complete PCM16 WAV without sending audio to any server. The reverb layer keeps each input sample and adds three delayed copies at 29.7, 37.1, and 41.1 milliseconds with disclosed weights of 0.50, 0.35, and 0.25 multiplied by the chosen amount. Because every step runs inside the current browser tab, no account is required, no plugin is installed, and the original file never leaves the device. The exported file is a fresh interleaved little-endian RIFF/WAVE in 16-bit signed PCM, sized to fit the input plus the entire reserved reverb tail rather than truncated mid-effect. The page reports channel count, sample rate, input and output frame counts, output peak, and exact WAV byte size after processing, so you can confirm the reverb pass actually changed the file before you save it. That is the entire contract: one file in, one deterministic reverb pass, one PCM16 WAV out.
That definition matters because most search results for "app to add reverb to audio" point at multi-effect digital audio workstations, mobile plugins, or cloud renderers that demand an account or a connection. A browser app keeps the workflow shorter: open the page, drop in one audio file, pick the reverb effect, set a percentage, and save the WAV. The trade-off is also explicit. This reverb is a lightweight three-tap reflection design, not a measured concert hall, plate, spring, studio, or branded hardware emulation, and it does not model room dimensions, frequency-dependent decay, stereo diffusion, damping, pre-delay, or convolution. Knowing that boundary up front is what lets you decide when the Audio Effects Online tool is the right pick and when it is not.
Why a browser app fits a quick reverb pass
A local browser app removes the friction of a full digital audio workstation when you only need one reverb pass on one file. There is nothing to install, no driver to update, and no account to create. The page itself reads the file through the browser's Web Audio implementation, decodes it into an AudioBuffer, applies the effect to every sample frame, and writes a new WAV. The W3C Web Audio API defines how the browser exposes channel count, sample rate, and frame length for that buffer, which is why the page can report those values honestly after processing (W3C Web Audio API 1.1).
The privacy story is also straightforward: file reading, decoding, sample processing, encoding, and download creation all happen in the current browser tab. There is no server render and no hidden AI transformation. Audio is not uploaded to Lizely or stored in an account. That makes the app practical for voice takes, dialogue drafts, podcasts, and rough sound-design sketches where you want a quick spatial lift without committing to a heavier plugin chain.
The reverb effect: three taps, three weights, one amount control
The reverb is a deterministic three-tap early-reflection design. Each tap is a delayed copy of the dry signal summed back into the output:
| Tap | Delay | Weight at 100% amount | Effective weight at 60% amount |
|---|---|---|---|
| 1 | 29.7 ms | 0.50 | 0.30 |
| 2 | 37.1 ms | 0.35 | 0.21 |
| 3 | 41.1 ms | 0.25 | 0.15 |
| Sum | — | 1.10 | 0.66 |
The amount control scales each tap linearly. At 0% the reverb is silent and you get the dry signal back. At 100% the listed weights apply directly. The complete delayed tail is allocated in advance, so the last reflection is not cut off at the end of the file.
Because the taps overlap with the dry signal, peak amplitudes can climb past 1.0. Consider a worked example: at 100% amount, the three taps add 0.50 + 0.35 + 0.25 = 1.10 extra copies of the dry peak when those delays overlap. If the dry peak sits at 0.95, the overlapping reverb peaks can reach 0.95 + (0.95 × 1.10) = 1.995. The tool does not hide that risk. It clips every finite output sample to the legal -1.0 through +1.0 range before quantizing to 16-bit PCM, so the WAV stays valid but the loudest combined samples lose information.
How to add reverb to audio with the app
The tool walks you through three steps. Each one is enforced by the page so the result is predictable.
- Choose one browser-decodable audio file no larger than 50 MiB. The page reads the file through your browser, so it must be a format the browser can decode. Names such as MP3, WAV, M4A, AAC, Ogg, WebM, and FLAC are accepted, but actual codec and profile support varies by browser and operating system. A recognized filename does not guarantee that a particular encoded file will decode; if decoding fails, the tool creates no result.
- Select the lightweight reverb effect and set a whole-number amount from 0% through 100%. The reverb tab is one of six available effects. The amount scales each tap weight linearly, so 50% gives half-strength reflections and 100% gives the full 0.50, 0.35, 0.25 set.
- Apply the effect locally, inspect frame counts and peak, then download the PCM16 WAV while keeping the untouched original. After processing, the page reports channel count, sample rate, input and output frame counts, output peak, and the exact WAV byte size. Download the new file and keep your source untouched so you can re-run with a different amount.
If your file is larger than 50 MiB, the run is refused before decoding. If the decoded result would exceed the 30 million channel-sample budget, the tool rejects the job without truncation and produces no file. Changing the file, the effect, or the amount invalidates older work in the same session, so stale jobs cannot replace a fresh result.
File inputs, output format, and the hard limits
The output is a fresh interleaved little-endian PCM16 RIFF/WAVE. Microsoft documents the WAV block alignment as channels × bits per sample ÷ 8, which at sixteen bits gives two bytes per mono frame and four bytes per stereo frame (Microsoft WAVEFORMATEX structure). The download preserves only the processed channel samples and sample rate. The source codec, bitrate, compression quality, tags, artwork, chapters, cue points, loop markers, and other container metadata are not preserved.
Before you start, confirm the file clears every limit that can stop the run:
- File size: one encoded file may be at most 50 MiB.
- Channel count: 1 to 8 equal-length channels after decoding.
- Sample rate: 8,000 to 192,000 Hz.
- Effect amount: an integer from 0 through 100.
- Channel-sample budget: decoded input and complete output must each stay within 30 million channel samples. Echo and reverb can pass the input budget yet fail the output budget because their tails add frames; in that case nothing is truncated and no file is created.
All limits are checked up front. The page also closes decode contexts, prevents stale jobs from replacing newer state, and revokes temporary object URLs on replacement and unmount. For a file-format specific walk-through with MP3 sources, the Add Reverb to an MP3 File Online Locally guide follows the same three steps against an MP3 source.
Other effects in the same app for related tweaks
Reverb is one of six bounded effects. Each one applies a single deterministic transformation per run, so to build a chain you download a result, inspect it, and feed the new WAV back in as a later input.
| Effect | What it does to every sample | Amount control | Best for |
|---|---|---|---|
| Echo | Keeps the input sample and adds a half-strength copy 200 ms later. | 0–100% scales the delayed copy linearly. | Slap-back repeats on a vocal or guitar take. |
| Lightweight reverb | Adds three early reflections at 29.7, 37.1, 41.1 ms with weights 0.50, 0.35, 0.25. | 0–100% scales each tap linearly. | Quick spatial lift on dialogue or a sound-design draft. |
| Reverse | Reads sample frames in opposite order and blends with the original. | 0–100% blends dry and reversed signal. | Hidden-message effects, transitional hits, risers. |
| Fade in | Scales from the selected dry proportion at the start to full level at the end. | 0–100% sets the starting dry level. | Building a song intro or smoothing a hard edit. |
| Fade out | Inverse of fade in: full level at the start, selected dry proportion at the end. | 0–100% sets the ending dry level. | Ending a podcast or track without an abrupt cut. |
| Peak normalization | Finds the largest absolute decoded sample and scales toward a 0.95 peak. | Not applicable; runs at the declared target. | Recovering headroom before a final limiter pass. |
Normalization is not loudness matching. It does not target LUFS, perceived volume, broadcast compliance, dynamics, true peak between samples, or equal loudness across recordings. Silent input stays silent. If you need a measured broadcast target, run a dedicated loudness meter on the exported WAV. If you want a full tour of how the six effects differ from each other, the Add Reverb and Echo to Audio Online: Six Local Effects guide covers the same tool from that angle.