Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Noisy speech recognition using de-noised multiresolution analysis acoustic features.

C P Chan1, P C Ching, T Lee

  • 1Department of Electronic Engineering, The Chinese University of Hong Kong, Shatin, New Territories, People's Republic of China.

The Journal of the Acoustical Society of America
|January 5, 2002
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Community burden of hepatitis A infection and risk of transmission in Hong Kong.

Hong Kong medical journal = Xianggang yi xue za zhi·2023
Same author

Adherence of nurses to annual seasonal influenza vaccination over a 5-year period.

The Journal of hospital infection·2021
Same author

Inhibition of RIG-I-dependent innate immunity by herpes simplex virus type I Us11 protein.

Hong Kong medical journal = Xianggang yi xue za zhi·2018
Same author

An unusual cause of acromegaly.

Hong Kong medical journal = Xianggang yi xue za zhi·2014
Same author

Should lidocaine spray be used to ease nasogastric tube insertion? A double-blind, randomised controlled trial.

Hong Kong medical journal = Xianggang yi xue za zhi·2010
Same author

Prostaglandin F(2alpha) stimulates MEK-ERK signalling but decreases the expression of alkaline phosphatase in dental pulp cells.

International endodontic journal·2010
Same journal

High-resolution depth estimation for multiple wideband sources in deep sea via sparse Bayesian learninga).

The Journal of the Acoustical Society of America·2026
Same journal

Depression markers in speech: An approach based on tract variables dynamics.

The Journal of the Acoustical Society of America·2026
Same journal

The oyster toadfish (Opsanus tau) alters active and diurnal calling amid vessel noise in New York City.

The Journal of the Acoustical Society of America·2026
Same journal

Experimental noise characterisation of phase-locked tandem-rotor in edgewise flight.

The Journal of the Acoustical Society of America·2026
Same journal

The tune-text-temporal synergy: Prosodic effects of final segmental weakening in Neapolitan.

The Journal of the Acoustical Society of America·2026
Same journal

Monitoring vessel movement above critical offshore infrastructure using distributed acoustic sensing.

The Journal of the Acoustical Society of America·2026
See all related articles

This study introduces multiresolution analysis (MRA) for robust speech recognition, enhancing de-noising capabilities. MRA features show improved phone recognition accuracy in noisy conditions compared to traditional methods.

Area of Science:

  • Signal Processing
  • Speech Recognition
  • Acoustics

Background:

  • Robust speech recognition is crucial for human-computer interaction.
  • Traditional acoustic features like MFCCs can degrade significantly in noisy environments.
  • De-noising techniques are essential for improving speech recognition performance.

Purpose of the Study:

  • To propose a novel application of multiresolution analysis (MRA) for extracting de-noising acoustic features.
  • To enhance the robustness of speech recognition systems against background noise.
  • To improve the prominence and contrast of consonant features.

Main Methods:

  • Constructing a mel-scaled wavelet packet filter-bank using MRA.
  • Computing subband powers as feature parameters for speech recognition.

Related Experiment Videos

  • Applying Wiener filtering to selected subbands with noise reduction for high-frequency bands.

    • Main Results:

    • Achieved a 32% phone recognition rate on the TIMIT database with 10-dB SNR white noise.
    • Demonstrated a noticeable improvement over Mel-Frequency Cepstral Coefficients (MFCC) with (29%) and without (20%) Cepstral Mean Normalization (CMN).
    • MRA features exhibited smaller distortion compared to clean speech, indicating effective de-noising.

    Conclusions:

    • Multiresolution Analysis (MRA) offers a promising approach for robust speech recognition by incorporating de-noising capabilities.
    • The proposed MRA-based features outperform standard MFCCs in noisy conditions.
    • The method effectively enhances consonant clarity and reduces feature distortion.