Download

1.0.1

Features

  • Automatic dynamic equalizer audio plugin (VST3 & AU)
  • For podcasting, streaming, events, broadcasting, game, film & TV sound and audio drama
  • Live and post-production
  • Voice mode with harmonics balancing EQ
  • FFT modes with spectral processing
  • High and low resolution options
  • 5-band parametric post & sidechain EQ
  • Real-time FFT spectrum display
  • External sidechain support

atkAutoEq is an automatic dynamic equalizer. FFT modes apply spectral processing, useful for taming resonances and evening out frequency response across a wide range of material. Voice mode uses spectral analysis to identify harmonics and applies equalization to normalize their levels, producing a natural and consistent tonal balance for speech. A built-in de-esser, and 5-band post EQ and sidechain EQ allow further manual shaping.

Parameters

Processing Mode

  • Mode: Selects processing algorithm.
    • Voice: Analyzes spectrum and normalizes harmonic levels using regular zero-latency IIR EQ processing. Best for speech.
    • FFT: Spectral processing. Higher resolution, higher latency.
    • FFT Low-Res: Spectral processing. Lower resolution, minimal CPU and latency.

Core Dynamics

  • Threshold: Detection threshold for gain control. Frequencies below this level are not processed.
  • Amount: Maximum gain compensation depth.
  • Sensitivity: Controls how aggressively frequency content is detected and processed.
  • Attack: Envelope attack time for gain changes.

Voice Mode

  • Hold: Freezes current EQ curve.
  • Reset: Resets all harmonic gains to zero.
  • Mix Gate: Gate for dry/wet routing. When signal is above threshold, wet processing is selected; otherwise dry signal is selected.
  • HPF: Enables 24 dB/oct high-pass filter.

De-Esser (Voice Mode)

  • Enabled: Enables sibilance reduction.
  • Threshold: Sibilance detection threshold.
  • Amount: Sibilance compression depth.

FFT Mode

  • FFT Attack: Envelope attack time.
  • FFT Release: Envelope release time.
  • FFT Smooth: Spectral smoothing amount. Higher values blur detail.
  • FFT Knee: Soft knee for compression curve.
  • FFT Mode: Selects normal or average-deviation ballistics mode.
  • Alt: Alternative “averaging” processing mode.

Processing Options

  • Expander: Inverts gain curve. Boosts frequencies instead of cutting.
  • Delta: Monitors I/O difference instead of output.
  • Mid/Side: Enables mid/side stereo processing.
  • External Sidechain: Uses external sidechain input for analysis and detection.
  • Post SC Det: Applies sidechain EQ before harmonic analysis.
  • Q: Spectral smoothing. Lower values smooth more across frequency.

Output

  • Pre-Gain: Input gain before processing.
  • Gain: Output makeup gain.
  • Dry/Wet: Blend between unprocessed and processed signal. Latency-compensated.
  • Channel Balance: Balance between left and right channels.

Post EQ (5 Bands)

Manual EQ applied after dynamic processing. Each band has frequency, gain, Q, and type controls.

  • Band 1: Peak, Low Shelf, or High-Pass (24 dB/oct).
  • Band 2: Peak or Low Shelf.
  • Band 3: Peak.
  • Band 4: Peak or High Shelf.
  • Band 5: Peak, High Shelf, or Low-Pass (24 dB/oct).

Sidechain EQ (5 Bands)

Separate EQ applied to sidechain detection path. Same band structure as post EQ. Shapes what the detector responds to without directly affecting the output signal.

Usage

Voice Mode

  1. Select Voice mode
  2. Adjust Threshold so processing engages only on speech, not background noise
  3. Set Amount to control how much EQ is applied
  4. Adjust Sensitivity for detection reliability
  5. Use HPF to clean up low-end rumble relative to detected fundamental
  6. Enable De-Esser and adjust threshold to catch problematic sibilance
  7. Shape final tone with Post EQ bands

FFT Mode

  1. Select FFT or FFT Low-Res mode
  2. Set Threshold to define the floor of processing
  3. Adjust Amount to control processing depth
  4. Use FFT Attack/Release to set response speed
  5. Adjust FFT Smooth to control spectral resolution smoothing

Notes/Tips

  • Voice mode works best on single-voice sources. Multi-speaker or music content is better suited for FFT modes.
  • Use sidechain EQ to make the detector emphasize or de-emphasize specific frequency ranges. E.g. If voice output sounds too bassy, boost low frequencies in sidechain EQ to compensate.
  • Delta mode is useful for monitoring what the processor is changing.