Audio Equalizer PRO

About Audio Equalizer PRO

This professional online audio equalizer lets you adjust frequency bands with precision. Choose from 3-band to 31-band EQ for detailed control. Features include real-time spectrum analyzer, stereo peak meters, high/low pass filters, custom preset management, and export to multiple formats including MP3 and FLAC. All processing happens locally in your browser using Web Audio API.

What does an audio equalizer actually do to my sound?

An equalizer (EQ) boosts or cuts the loudness of specific frequency bands inside an audio signal without re-recording it. Human hearing spans roughly 20 Hz to 20 kHz, and every instrument, voice, or noise source occupies its own region of that range. By raising the gain at 80 Hz you make a kick drum thump harder; by cutting 3 kHz you take harshness out of a vocal. This tool applies digital filters in real time, so the original file is untouched until you export. EQ is non-destructive math: each band multiplies the signal by a frequency-dependent gain, and the inverse curve can usually restore the original. It is the single most-used corrective and creative process in mixing, mastering, podcast post, hearing aids, and live sound.

Why does cutting frequencies usually sound better than boosting them?

Subtractive EQ — pulling problem bands down — preserves headroom and avoids the phase distortion and noise that aggressive boosts introduce. A 6 dB boost doubles the amplitude of that band, pushing peaks closer to clipping and amplifying any noise sitting in the same range. A 6 dB cut, by contrast, attenuates whatever is annoying without adding energy elsewhere. Mastering engineers follow the maxim "cut narrow, boost wide": surgically notch out a resonance with a high-Q cut, but use broad, gentle bell boosts when sweetening. If you find yourself boosting more than 3 dB across multiple bands, the underlying mix balance is probably wrong and EQ alone cannot fix it. Always A/B against bypass at matched loudness.

How do I find the right frequency to cut or boost?

Use the "sweep and find" technique: temporarily set a band to a narrow Q (around 6-10) with a +10 dB boost, then slowly drag the centre frequency across the spectrum while the audio plays. The offending resonance or sweet spot will jump out as the boost amplifies it. Once located, flip the gain to a cut (for problems) or reduce the boost to a sensible 2-4 dB (for enhancement) and widen the Q so it sounds musical. Common sweep targets: 200-400 Hz for muddiness, 1-2 kHz for boxy honk, 2-5 kHz for harshness, 6-10 kHz for sibilance, and 60-120 Hz for low-end weight. Always solo the EQ band briefly to confirm, then evaluate in the full mix.

What is the difference between graphic, parametric, and dynamic EQ?

Graphic EQs offer a fixed set of bands (commonly 10 or 31 in ISO standard third-octave spacing) each with a single slider for gain — fast for live sound and consumer use but inflexible. Parametric EQs let you choose centre frequency, gain, and Q (bandwidth) per band, so each filter can be surgical or broad as needed; this is the studio standard for mixing and mastering. Dynamic EQs combine parametric with a compressor: the gain change only engages when the band exceeds a threshold, so you can tame a vocal that occasionally gets honky at 1.5 kHz without dulling it the rest of the time. Linear-phase EQs preserve transient timing at the cost of pre-ringing and latency, useful for mastering but risky on drums.

What does Q (bandwidth) mean and how do I set it?

Q is the ratio of centre frequency to bandwidth, so a higher Q means a narrower, more surgical filter. A Q of 0.7 covers roughly two octaves and sounds broad and musical — great for gentle tonal shaping. A Q of 1.4 is about one octave, useful for general corrections. Q values of 4-10 are surgical, ideal for notching out a single resonant frequency such as a mic-stand rumble or a snare ring. Above Q 20 you are essentially doing notch filtering for hum (50/60 Hz) or whistle removal. The trade-off: narrow Q changes phase sharply around its centre, which can make drum transients sound "smeared," so reserve high Q for problem-solving and use low Q for tone shaping.

What are high-pass and low-pass filters and when should I use them?

A high-pass filter (HPF, also called low-cut) removes frequencies below its cutoff while letting high frequencies pass — use it to roll off rumble, AC noise, traffic, and unnecessary subsonics. A typical podcast vocal benefits from a 12-24 dB/octave HPF at 80-120 Hz; an acoustic guitar can usually go up to 100 Hz with no loss. Low-pass filters (LPF, high-cut) remove frequencies above their cutoff, useful for taming hiss, cymbal bleed, or reproducing vintage telephone effects (band-pass 300 Hz to 3 kHz). Filter slope is measured in dB per octave: 6, 12, 18, 24, 36, and 48 are common. Steeper slopes remove more aggressively but add more phase shift and pre-ringing near the cutoff.

What is linear-phase EQ and when is it worth the latency cost?

Standard (minimum-phase) EQs introduce frequency-dependent phase shifts: different frequencies are delayed by different amounts, which can smear transients and shift the perceived attack of percussion. Linear-phase EQs use FIR filters that delay all frequencies by the same amount, preserving the original waveform shape — but at the cost of added latency (often 20-100 ms) and pre-ringing (a subtle ghost before each transient). Use linear-phase on the mix bus or master where transient integrity matters, on parallel-processed tracks where phase cancellation with the dry signal is a concern, and on stereo elements where channel-symmetric processing is critical. Avoid it on tight drums, kick, and snare where the pre-ring becomes audible. For tracking and live monitoring, stick with minimum-phase to keep latency negligible.

How does EQ interact with compression and the rest of the signal chain?

Order matters. EQ before compression shapes what the compressor reacts to: a high-pass before the comp prevents bass from triggering excessive gain reduction; a boost at 5 kHz makes the comp clamp harder on sibilance. EQ after compression shapes the already-tamed signal cosmetically, often with broader bell curves for tone. A common pro chain is HPF -> subtractive EQ (problem removal) -> compressor -> additive EQ (tone shaping) -> saturator -> limiter. On the master bus, EQ before the limiter prevents specific frequencies from forcing the limiter to pump on the whole mix. Always re-evaluate gain staging after EQ changes: a 6 dB boost at 100 Hz can add 3-4 dB to overall RMS even if peak meters look similar, so adjust output to compensate before A/B testing.

How do I check loudness and true-peak compliance for streaming or broadcast?

Use the Loudness & True-Peak panel. It renders your EQ'd audio and measures Integrated Loudness in LUFS using ITU-R BS.1770 K-weighting (a high-shelf plus high-pass model of the ear) with EBU R128 gating: an absolute gate at -70 LUFS and a relative gate at -10 LU below the ungated mean. It also reports Loudness Range (LRA) and 4x-oversampled true-peak (dBTP), which catches inter-sample overs that a normal sample peak meter misses — these are common after an EQ boost and routinely read +0.5 to +1 dBTP higher than the sample peak. Pick a target (Spotify/YouTube ~-14 LUFS, EBU broadcast -23 LUFS, Apple Podcasts -16 LUFS, all with a -1 dBTP ceiling) and the pass/fail badge tells you if your deliverable is within +/-1 LU of target and under the true-peak limit before you export.

What bit depth and sample rate does the export use?

WAV export is 16-bit PCM and preserves the source sample rate (44.1 kHz, 48 kHz, 96 kHz, etc.) — the rendering engine never resamples your audio, so timing and pitch are untouched. MP3, OGG, and FLAC are encoded from that render via FFmpeg in your browser. If you need 24-bit or 32-bit float masters for further mastering, render to WAV here and convert in a DAW; 24-bit matters most when you will apply additional gain or processing downstream, because the extra headroom keeps the noise floor below audibility. For final delivery to streaming platforms, 16-bit WAV or a high-bitrate lossy file at the platform's target loudness is normally sufficient.