Automatic Speaker Recognition

Created by Group 3 of the USTH Digital Signal Processing 2020 course, this dataset supports automatic speaker recognition tasks using audio recordings. It was last updated in July 2023 by author dinhanhx and focuses on human voice signal processing.

Use Cases

• Extracting MFCC features from .wav files for acoustic analysis
• Training Gaussian Mixture Models (GMM) for speaker identification using sklearn
• Testing digital signal processing algorithms on human voice recordings

Strengths

• Released under a permissive MIT license
• Specifically tagged for MFCC feature extraction and GMM modeling
• Curated for digital signal processing (DSP) academic standards

Limitations

• Unknown sample size and speaker diversity
• Lack of explicit metadata or label files in the provided description
• Potential geographic bias as it originated from a specific university project in 2020

Provenance

Source

University of Science and Technology of Hanoi (USTH) Digital Signal Processing 2020 Group 3

Time Range

2020

Freshness

Last updated July 2023; no ongoing maintenance is indicated.