Audio2Pitch¶

standard mode | Pitch category

Inputs¶

frame (vector_real) - the input frame to analyse

Outputs¶

pitch (real) - detected pitch in Hz

pitchConfidence (real) - confidence of detected pitch (from 0.0 to 1.0)

loudness (real) - detected loudness in decibels

voiced (integer) - voiced frame categorization, 1 for voiced and 0 for unvoiced frame

Parameters¶

frameSize (integer ∈ [1, ∞), default = 1024) :
size of input frame in samples

loudnessThreshold (real ∈ [-∞, 0], default = -51) :
loudness level above/below which note ON/OFF start to be considered, in decibels

maxFrequency (real ∈ [10, 20000], default = 2300) :
maximum frequency to detect in Hz

minFrequency (real ∈ [10, 20000], default = 60) :
minimum frequency to detect in Hz

pitchAlgorithm (string ∈ {pitchyin, pitchyinfft}, default = pitchyinfft) :
pitch algorithm to use

pitchConfidenceThreshold (real ∈ [0, 1], default = 0.25) :
level of pitch confidence above/below which note ON/OFF start to be considered

sampleRate (integer ∈ [8000, ∞), default = 44100) :
sample rate of incoming audio frames

tolerance (real ∈ [0, 1], default = 1) :
sets tolerance for peak detection on pitch algorithm

weighting (string ∈ {custom, A, B, C, D, Z}, default = custom) :
string to assign a weighting function

Description¶

This algorithm computes pitch with various pitch algorithms, specifically targeted for real-time pitch detection on audio signals. The algorithm internally uses pitch estimation with PitchYin (pitchyin) and PitchYinFFT (pitchyinfft).

Source code¶

C++ source code

C++ header file