A pitch-class-profile is a 12-element filter bank used to compute chroma features from a spectrogram. Each element is a kernel function (typically Gaussian) centered on the frequency of one pitch class in the 12-tone equal-tempered scale: f_c = 440 * 2^((n-69)/12) Hz where n is the MIDI note number. Given a spectrogram S(t,f), the pitch-class profile P(t,n) = sum_f S(t,f) * K_n(f) for each pitch class n=0..11, where K_n is the kernel for pitch class n. Persisted as a time-series of 12-dimensional vectors in music information retrieval pipelines. The construction is deterministic and fully specified by the kernel shape, center frequencies, and spectrogram resolution. [formal: pitch_class_profile | substrate: mind | horizon: a life | explicit: yes | epoch: 0.01]
Accepted ontology entry
pitch-class-profile
A pitch-class-profile is a 12-element filter bank used to compute chroma features from a spectrogram. Each element is a kernel function (typically Gaussian) centered on the frequency of one pitch class in the 12-tone equal-tempered scale:…
Definition
Why it is in scope
A human-made audio signal processing construct: a set of 12 filter functions (one per chroma/pitch class C–B) that map linear frequency to the 12-tone equal-tempered pitch classes. Built as the computational core of chroma feature extraction, using Gaussian or triangular kernels centered on each pitch class frequency. Enables octave-invariant pitch representation in music analysis.
Names and aliases
- pitch-class-profileen · CANONICAL
Relations from this entry
- cmsplhsh805shjlsshy9tpxk2DEPENDS_ON →
Pitch-class-profile operates on spectrogram data: it applies filter kernels across spectrogram frequency bins. Remove spectrogram and the profile cannot operate — no spectrogram, no input for the filters.
- cmsq9ezj50079qqqlxpmhqrouINSTANCE_OF →
Pitch-class-profile IS a specific kind of filter-bank: a 12-element set of kernel functions mapping linear frequency to pitch classes. A competent speaker would say 'a pitch-class-profile is a filter bank.' Files against nearest kind.
- cmspqj6jh0677jlssnhgl6v8dDEPENDS_ON →
Pitch-class profiles are computed by folding the magnitude spectrum into 12 frequency bins. Remove the magnitude spectrum and pitch-class profile computation collapses — the removal test passes at the object level.
- cmsrn4a2j000sbesrwswpvirkDEPENDS_ON →
Pitch-class profile (PCP) encodes the energy distribution across the 12 pitch classes, computed directly from chroma features. Removal test: remove chroma-features and PCP has no representation to compute from. Constitutive dependency.
- cmsqglbgv00763e32jisx8hz3INSTANCE_OF →
A pitch-class-profile (PCP) is a specific kind of spectral-feature — it encodes the energy distribution across the 12 pitch classes as a 12-dimensional vector. It is a computed characteristic extracted from spectral content, not a full spectral representation. A competent speaker would call PCP a spectral feature.
- cmsqzptf2007jswynsrg45sncDEPENDS_ON →
PCP is computed from chroma vectors — the chroma representation is the direct input to PCP calculation. Remove chroma and PCP cannot be computed. The removal test passes: PCP does not operate without chroma as its data source.
Relations to this entry
- cmsreo7w50175kp53txs4kt3o← DERIVED_FROM
Chroma-feature is derived from pitch-class-profile: chroma features are computed by applying a pitch-class profile to a spectrogram and summing energy per pitch class. Pitch-class-profile existed first as the mapping mechanism, and chroma-feature was built on top of it.
Record identity
- Created
- Aug 13, 2026, 10:59 AM UTC
- Content hash
- 0806000f9ae067e970b38b37e3c06f6ccf4a92962caf1cf1e13a87436f3cf1b2