A feature extraction method and system for speech signals

By adaptively adjusting the window size, the problem of inappropriate window size selection in speech signal processing is solved, enabling accurate extraction of speech signal features and improving the accuracy of speech recognition and speaker identification.

CN121922150BActive Publication Date: 2026-06-05GUANGZHOU JIUSI INTELLIGENT TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
GUANGZHOU JIUSI INTELLIGENT TECH CO LTD
Filing Date
2026-03-26
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

Existing technologies lack the ability to adaptively adjust window size in speech signal processing, making it difficult for spectrograms to accurately reproduce the true features of speech signals, thus affecting the accuracy of speech recognition and speaker identification.

Method used

An adaptive window size adjustment method is adopted. By calculating the stability limit threshold and the stability of spectral changes, the window size is dynamically adjusted to extract speech signal features. This includes calculating the stability and the complexity of the patterns, and selecting a suitable window size for time-frequency analysis.

Benefits of technology

It achieves accurate extraction of frequency domain information of speech signals while preventing insufficient samples or excessive introduction of surrounding data, thus improving the accuracy and stability of speech signal feature extraction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121922150B_ABST
    Figure CN121922150B_ABST
Patent Text Reader

Abstract

The present application relates to the field of data processing, and particularly relates to a feature extraction method and system of a voice signal. The method comprises the steps of: obtaining a voice signal; for any data in the voice signal, obtaining a plurality of data segments with different lengths centered on the data, calculating a stable definition threshold, starting from a reference length, continuously increasing the window size, after any time of window size adjustment, obtaining a window centered on the data based on the adjusted window size, performing Fourier transform on the signal in the window to obtain a frequency spectrum, calculating a stability degree, if the stability degree is less than the stable definition threshold for the first time, taking the window size after this adjustment as a target window size; performing short-time Fourier transform on the voice signal based on the target window size of each data in the voice signal to obtain a spectrogram, and realizing feature extraction of the voice signal based on the spectrogram. The accuracy of the extracted spectrogram is improved by setting a suitable window size.
Need to check novelty before this filing date? Find Prior Art