Analyzing long-term rhythm variations in Mising and Assamese using frequency domain correlates

TL;DR

This paper investigates long-term speech rhythm variations in Mising and Assamese using frequency domain analysis of LF spectrograms, proposing a novel feature extraction approach for classifying these low-resourced languages.

Contribution

It introduces a new methodology for analyzing speech rhythm variations in low-resource languages through frequency domain features without requiring prior annotation.

Findings

Effective classification of Mising and Assamese based on rhythm features

Demonstrates the utility of rhythm formant analysis in low-resource language classification

Shows that LF spectrogram features can distinguish language rhythm differences

Abstract

The current work explores long-term speech rhythm variations to classify Mising and Assamese, two low-resourced languages from Assam, Northeast India. We study the temporal information of speech rhythm embedded in low-frequency (LF) spectrograms derived from amplitude (AM) and frequency modulation (FM) envelopes. This quantitative frequency domain analysis of rhythm is supported by the idea of rhythm formant analysis (RFA), originally proposed by Gibbon [1]. We attempt to make the investigation by extracting features derived from trajectories of first six rhythm formants along with two-dimensional discrete cosine transform-based characterizations of the AM and FM LF spectrograms. The derived features are fed as input to a machine learning tool to contrast rhythms of Assamese and Mising. In this way, an improved methodology for empirically investigating rhythm variation structure without prior annotation of the larger unit of the speech signal is illustrated for two low-resourced languages of Northeast India.