Audio Pitch Shifter

Pitch shifter with three selectable algorithms: Granular (overlap-add), Phase Vocoder (FFT-based), and Delay Crossfade (4-tap delay line, H910 style).

The pitch shifter transposes the incoming audio signal up or down by a specified interval in semitones. Three algorithms offer different trade-offs between quality, latency, and CPU usage:

• Granular: splits the signal into overlapping grains and replays them at a different rate. Good general-purpose quality.
• Phase Vocoder: uses FFT analysis/resynthesis to shift frequency bins. Best quality for small intervals but introduces latency equal to the FFT size.
• Delay Crossfade: classic hardware-style pitch shifting using four cross-fading delay taps (Eventide H910 approach). Lowest latency and CPU, but may produce audible artifacts on large intervals.

Settings

bypass

Bypasses the module processing when activated (ON).

• OFF (0): Module processes normally
• ON (1): Input passes directly to output without processing

in

Audio input flow(s).

out

Audio output flow(s).

pitch

Pitch shift amount in semitones. Negative values shift down, positive values shift up. At 0 the signal passes through unmodified.

mix

Dry/wet blend. At 0 the output is the unprocessed input signal; at 1 the output is fully pitch-shifted. Intermediate values blend both for parallel processing effects.

mode

Selects the pitch shifting algorithm:

• Granular — time-domain overlap-add. Good general quality, moderate latency (~40 ms depending on grain size). Best for moderate shifts (up to ±12 semitones).
• Phase Vocoder — FFT-based spectral processing. Highest quality for small shifts, but introduces latency proportional to the FFT size. Best for ±7 semitones or less.
• Delay Crossfade — 4-tap delay line with Blackman crossfade (H910 style). Near-zero latency, minimal CPU. Works well for ±12 semitones but produces more artifacts than the other modes.

buffer size

Grain size in milliseconds for Granular mode, or crossfade window size for Delay Crossfade mode.

• Smaller grains (10-20 ms): tighter transients, more metallic artifacts.
• Larger grains (60-100 ms): smoother sound, more time smearing.
• Not used in Phase Vocoder mode (use fft size instead).

fft size

FFT analysis window size for Phase Vocoder mode. Larger values give better frequency resolution (fewer artifacts) but higher latency.

• 512 — ~10 ms latency, lowest quality
• 1024 — ~21 ms latency
• 2048 — ~42 ms latency (default, good balance)
• 4096 — ~85 ms latency, highest quality

Not used in Granular or Delay Crossfade modes.

See also

• Audio Disto Bit Crusher
• Audio Disto Wave Shaper
• Audio Disto Foldback
• Audio Disto Rectifier
• Audio Disto Diode Clipper
• Audio Disto Tape Saturation
• Audio Disto Tube Saturation
• Effects
• Audio Flanger
• Audio Disto Bit Crusher
• Audio Phaser
• Granulator (USER MODULE)
• Audio Tape Echo
• Audio Chorus
• Audio Pitch Shifter
• Audio Delay Amplitude Time Full
• Audio Disto Wave Shaper
• Audio Disto Foldback
• Audio Disto Rectifier
• Audio Delay Amplitude Time
• Audio Rotary Speaker
• Audio Delay Cloud
• Audio Disto Diode Clipper
• Audio Octave Divider
• Audio Disto Tape Saturation
• Audio Envelope Filter
• Audio Spring Reverb
• Audio Disto Tube Saturation
• Audio Delay
• Audio Ring Modulator