U.S. flag An official website of the United States government.
Official websites use .gov

A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS

A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

i

Predicting the Intelligibility of Speech Corrupted by Nonlinear Distortion



Details

  • Personal Author:
  • Description:
    Common methods for predicting the intelligibility of speech, the speech transmission index (STI) and the speech intelligibility index (SII), fail when the speech signal is corrupted by nonlinear distortion, e.g., center clipping (Steeneken & Houtgast, 2002). This limitation restricts the applicability of the metrics to many digital communication systems, where peak and center clipping are often unwanted byproducts of the signal processing. For distorted speech, the performance of the SII may be improved by calculating the speech signal-to-'noise' (or distortion) ratio from the coherence for three amplitude ranges, and combining the results to assess the intelligibility (Kates and Arehart, 2005). In this paper we explore modifications to the STI to permit estimates of the intelligibility of speech corrupted within a communication system by center clipping. This type of distortion occurs when a signal rapidly changes polarity from a non-zero value, and is associated with quantization errors in digital signal processing. The development of models is first briefly described, followed by a summary of the psychophysical experiment employed to evaluate their performance. Results are presented for word intelligibility in speech-spectrum shaped noise, which is well predicted by the STI, and for speech subjected to center-clipping. [Description provided by NIOSH]
  • Subjects:
  • Keywords:
  • ISSN:
    0711-6659
  • Document Type:
  • Funding:
  • Genre:
  • Place as Subject:
  • CIO:
  • Topic:
  • Location:
  • Pages in Document:
    80-81
  • Volume:
    38
  • Issue:
    3
  • NIOSHTIC Number:
    nn:20058842
  • Citation:
    Can Acoust 2010 Sep; 38(3):80-81
  • Federal Fiscal Year:
    2010
  • NORA Priority Area:
  • Performing Organization:
    University of Connecticut School of Medicine and Denistry, Farmington, Connecticut
  • Peer Reviewed:
    False
  • Start Date:
    20060801
  • Source Full Name:
    Canadian Acoustics
  • End Date:
    20120731
  • Collection(s):
  • Main Document Checksum:
    urn:sha-512:1faff4d7459de66090d9e91f800f84a2a65380d43cd0199ed3652a4dc5bd91f85a72e216922c071a085bd93da3a7e20965bc18d3324e954d8c4fb0f734bf6f1d
  • Download URL:
  • File Type:
    Filetype[PDF - 154.90 KB ]
ON THIS PAGE

CDC STACKS serves as an archival repository of CDC-published products including scientific findings, journal articles, guidelines, recommendations, or other public health information authored or co-authored by CDC or funded partners.

As a repository, CDC STACKS retains documents in their original published format to ensure public access to scientific information.