Transchromagrams for Accurate Audio Tonality Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computational methods for extracting tonality information from audio signals are limited in accuracy and do not effectively capture the dynamic changes in musical notes over time, leading to unreliable harmony identification and key detection in music.
Innovation Solution
The use of transchromagrams, which are probabilistic note transition matrices derived from chromagrams, to characterize audio data by representing the likelihood of note transitions over time, allowing for the analysis and comparison of audio sequences to identify tonal similarities and differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional chromagram methods are used to extract tonality information, then the computational process is simple, but the accuracy of harmony identification and key detection is limited
Solution Approach 1:
The patent transitions from traditional chromagram representation to transchromagram by adding a temporal dimension through note transition probabilities. This dimensional expansion captures dynamic changes in musical notes over time, resolving the contradiction by improving measurement precision through enhanced representation while managing complexity through structured probabilistic modeling.
Solution Approach 2:
The invention changes the fundamental parameters of audio characterization by introducing note transition probabilities and temporal dynamics. This parameter transformation from static frequency mapping to dynamic transition modeling improves tonality extraction accuracy while maintaining computational feasibility through established probability theory frameworks.
2Loss of information
If static chromagram representation is used, then the data structure is simple, but the dynamic changes in musical notes over time are not captured
Solution Approach 1:
The patent adds a temporal dimension to the chromagram data structure by incorporating note transition probabilities across time frames. This dimensional enhancement preserves information about how musical notes evolve over time while organizing the complexity into a structured probabilistic framework that remains computationally manageable.
Solution Approach 2:
The invention transforms the static chromagram into a dynamic transchromagram that captures temporal variations in note transitions. This dynamic representation preserves information about musical evolution over time while using probabilistic models to manage the increased data complexity in a systematic way.
3Reliability
If traditional Fourier Transform or CQT is used, then the computational method is well-established, but the reliability of harmony identification is insufficient
Solution Approach 1:
The patent applies preliminary transformations by first converting audio signals to chromagrams using established methods (Fourier Transform or CQT), then further processing these into transchromagrams with note transition probabilities. This two-stage approach builds upon well-established computational foundations while adding the necessary complexity to improve harmony identification reliability.
Solution Approach 2:
The transchromagram acts as an intermediary data structure between traditional chromagram analysis and final harmony identification. This intermediate representation with note transition probabilities bridges the gap between simple spectral analysis and reliable harmony detection, managing computational complexity through structured probabilistic reasoning.
Data Source
AI summary
Methods, systems and apparatus to characterize audio using transchromagrams are disclosed. An example apparatus includes a transchromagram generator to generate a data structure based on a set of transition matrices corresponding to a plurality of time frames of audio data, the data structure indicative of probabilities that first musical notes will transition to second musical notes, a database controller to prompt a database to store the data structure within the audio data, and a notification manager to generate, based on a comparison between query audio data and the stored data structure of the audio data, a notification identifying at least one characteristic of the query audio data.