Club sound simulator: hear your mix on eight playback systems

Golden Ear renders your track through eight playback environments: warehouse, club, festival, car, laptop speaker, earbuds, vinyl and a mono Bluetooth speaker. Each environment uses a synthetic impulse response and tone shaping, and the tool lists the problems that are likely to show up there. The processing runs in your browser, and a free Unveil account is required.

Why a mix sounds different on a big system

Studio monitors and headphones give you a controlled picture of a mix. A club, a warehouse or a festival stage does not. Large PAs have strong sub-bass, rooms add long reverb tails, and the audience absorbs high frequencies. Something that sounds tight at your desk can turn into a boom or a blur once it is reinforced by a room.

You cannot fully predict this without hearing the mix on real systems, but you can test for the most common translation problems in advance.

What to check: low end, mono and highs

Sub energy. Make sure the kick and bass are clean below 100 Hz and not stacked at the same frequency. In large spaces, excess sub builds up and hides the rest of the mix.
Mono compatibility. Many club systems and small speakers fold the signal toward mono. Elements that rely on wide stereo effects can thin out or cancel, so check the mix in mono, especially the bass and the vocal.
Midrange presence. Reverb and reflections can smear the 1 to 3 kHz region where vocals and snares live. A mix that is too soft in the mids loses impact in a loud room.
High frequencies. Outdoor and crowded spaces absorb high frequencies, while earbuds and laptop speakers can emphasize the upper mids and make hi-hats harsh.
Small speakers. If the bass has no energy above 200 to 300 Hz, it disappears on a laptop or a phone. Harmonics let the bass line translate.

What the Golden Ear environments do

Each environment in the code has a profile: a reverb time, pre-delay, low shelf, midrange peak, high shelf, a low cut that stands in for the speaker's bass limit, and a wet/dry blend. For example, the warehouse uses a 2.8 second reverb time, the club 1.2 seconds, and the festival 0.7 seconds with a strong high-frequency reduction. The laptop profile has a strong low-frequency cut, and the mono Bluetooth profile cuts below 120 Hz.

The tool builds a synthetic impulse response from exponentially decaying noise, shapes it with those filters, and convolves your audio with it using an offline audio context. The result blends the original and the convolved signal. It then measures energy in six frequency bands, compares the rendered result with the source, and flags issues for the chosen environment.

Step-by-step workflow

Sign in with a free Unveil account and open the Golden Ear tool.
Choose a track. You can also try the built-in demo signal instead of your own file.
Pick an environment and press Simulate Playback. A real run costs 2 ProCoins.
Listen with the play button. Use the environment buttons to move between systems. Renders you have already made are cached.
Read the issues list. Each entry has a severity and a suggested fix. Make changes in your mix, bounce again, and test again.

Preparing your file

Use a full mix bounce that is representative of your final level, in a format your browser can decode. Test the unmastered mix as well as the master if you can, because translation problems are easier to fix before mastering. Longer files take longer to render, and the tool suggests trying a shorter or different file if processing fails.

Honest limits

The environments are simplified models. They do not know the real size of a room, the PA, the position of the listener, or the settings of a sound engineer. The time stamps shown next to each issue are illustrative labels, not measured positions in your track, so use the issue text and the listening itself rather than the clock. Issue detection compares band energy before and after processing, which is a rough guide. Treat every result as a prompt to listen again on a real system, not as a pass or fail verdict.

A common mistake is to fix everything the simulator flags. Prioritize issues you can also hear, and keep a reference track on hand so you can judge whether a change actually improved the mix.

What you get

Eight environments, each with its own reverb time, low-end, midrange and high-frequency profile.
Switch between environments during playback, with finished renders cached so repeat switches are quick.
A list of issues per environment, rated critical, warning or info, each with a suggested fix.
Audio is decoded and rendered in your browser with the Web Audio API, so the track is not uploaded to a server.

Questions

Are these recordings of real venues?

No. The impulse responses are synthetic, generated from decaying noise shaped with filters for each environment. The simulator is an approximation for spotting likely problems, not a measurement of any real room or sound system.

Does my audio leave my computer?

The tool states that convolution runs in your browser with the Web Audio API, and the processing code runs client-side. A free Unveil account is still needed to start a run.

What does a run cost?

Starting a simulation costs 2 ProCoins, and signed-out visitors are sent to sign-up first. The page also offers a demo that uses a built-in test signal instead of your file.

Analyze your mix — free