Skip to content

Cadence Detection

Extract and analyze predicted cadences in Renaissance counterpoint using CRIM Intervals.

Overview

The get_cadences tool uses machine learning to identify and classify cadences in Renaissance polyphonic music. It analyzes voice leading patterns and harmonic progressions to predict cadence locations and types.

Usage

from encoding_music_mcp.tools import get_cadences

result = get_cadences("Morley_1595_01_Go_ye_my_canzonettes.mei")

Parameters

Parameter Type Required Description
filename str Yes Name of the MEI file to analyze

Returns

Returns a dictionary with the following keys:

Key Type Description
filename str The input filename
cadences str CSV representation of the cadences dataframe

Output Format

The cadences dataframe includes the following columns:

Column Description
Composer Composer name from MEI metadata
Title Work title from MEI metadata
Measure Measure number where cadence occurs
Beat Beat position within the measure
Progress Progress through the piece (0.0-1.0)
CadType Cadence type (Authentic, Plagal, etc.)
Tone The tonal centre of the cadence
CVFs Cadential Voice Functions for each voice

Example Output

Composer,Title,Measure,Beat,Progress,CadType,Tone,CVFs
"Morley, Thomas",Go ye my canzonettes,14,2.0,0.186,Authentic,G,"['Ct', 'T']"
"Morley, Thomas",Go ye my canzonettes,21,4.0,0.298,Plagal,D,"['Ct', 'T']"
"Morley, Thomas",Go ye my canzonettes,30,4.0,0.421,Authentic,C,"['Ct', 'T']"

Cadence Types

The tool identifies several cadence types common in Renaissance music:

Authentic Cadence

  • Strongest conclusive cadence
  • Typically involves dominant-to-tonic motion
  • Features characteristic voice leading patterns

Plagal Cadence

  • Subdominant-to-tonic motion
  • Often called the "Amen" cadence
  • Provides a sense of resolution

Phrygian Cadence

  • Half cadence with distinctive semitone motion in the bass
  • Common in Renaissance music
  • Often signals a temporary resting point

Evaded Cadence

  • Expected cadence is avoided or interrupted
  • Creates harmonic surprise or continuation

Cadential Voice Functions (CVFs)

The CVFs column describes the role each voice plays in the cadence:

Function Description
T Tenor (structural voice)
Ct Cantus (often the highest voice)
B Bassus (lowest voice)
A Altus (middle voice)

Best Practices

Suitable Repertoire

This tool is optimised for: - Renaissance polyphony (15th-17th century) - Sacred and secular vocal music - Italian, English, and Franco-Flemish styles - Works with clear voice leading

Interpretation

Remember that: - Cadence detection uses machine learning predictions - Results may vary based on texture and style - Some cadences may be more clearly defined than others - Context and musical judgment are still important

Analysis Workflow

  1. Initial Detection: Run get_cadences to identify potential cadences
  2. Review Locations: Check measure and beat positions
  3. Verify Types: Confirm cadence types match musical context
  4. Analyse Patterns: Look for recurring cadential formulas
  5. Compare Pieces: Study cadence usage across works

Example Analysis

# Analyse cadences in a Morley canzonet
result = get_cadences("Morley_1595_01_Go_ye_my_canzonettes.mei")

# The output shows:
# - 8 cadences throughout the piece
# - Mix of Authentic and Plagal types
# - Cadences at structural points (measures 14, 21, 30, etc.)
# - Consistent voice functions in two-part texture

Use Cases

Form Analysis

  • Identify section boundaries
  • Map large-scale structure
  • Study phrase lengths

Style Comparison

  • Compare cadential practices between composers
  • Analyse historical trends
  • Study regional differences

Voice Leading

  • Examine cadential formulas
  • Study approach patterns
  • Analyse resolution types

Structural Analysis

  • Identify important arrival points
  • Understand tonal hierarchy
  • Map harmonic rhythm

Technical Details

The cadence detection algorithm: - Analyses simultaneous intervals between voices - Identifies characteristic voice-leading patterns - Considers melodic motion in approach to cadence - Classifies cadence type based on learned patterns - Extracts tonal centre from harmonic context

Limitations

  • Optimised for Renaissance counterpoint
  • May be less accurate for:
  • Highly chromatic passages
  • Unusual textures
  • Non-Western music
  • Instrumental music with different conventions
  • Predictions should be verified musically

See Also