Detail publikace

Time-Frequency Analysis for Voice Activity Detection

Originální název

Time-Frequency Analysis for Voice Activity Detection

Anglický název

Time-Frequency Analysis for Voice Activity Detection

Jazyk

en

Originální abstrakt

This paper introduces two different ways of time-frequency representations for voice activity detection (VAD). The first method is based on the chirp transformation of the signal, while the second method is based on wavelet decomposition. The performance of both methods with clean speech and speech in noisy conditions is evaluated, and their limitations are discussed.

Anglický abstrakt

This paper introduces two different ways of time-frequency representations for voice activity detection (VAD). The first method is based on the chirp transformation of the signal, while the second method is based on wavelet decomposition. The performance of both methods with clean speech and speech in noisy conditions is evaluated, and their limitations are discussed.

BibTex


@inproceedings{BUT18979,
  author="T. {Pham} and Marián {Képesi} and Gernot {Kubin} and Luis {Weruaga} and Milan {Sigmund} and Tomáš {Dostál}",
  title="Time-Frequency Analysis for Voice Activity Detection",
  annote="This paper introduces two different ways of time-frequency representations for voice activity detection (VAD). The first method is based on the chirp transformation of the signal, while the second method is based on wavelet decomposition. The performance of both methods with clean speech and speech in noisy conditions is evaluated, and their limitations are discussed.",
  address="M.H. Hamza",
  booktitle="Proceedings of IASTED International Multi-Conference on Applied Informatics, part Signal Processing, Pattern Recognition, and Applications",
  chapter="18979",
  institution="M.H. Hamza",
  year="2006",
  month="february",
  pages="244",
  publisher="M.H. Hamza",
  type="conference paper"
}