The contribution of changes in F0 and spectral tilt to increased intelligibility of speech produced in noise

Abstract

To cite this version:Youyi Lu, Martin Cooke. The contribution of changes in F0 and spectral tilt to increased intelligibility of speech produced in noise. Speech Communication, Elsevier : North-Holland, 2009, 51 (12) This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. ACCEPTED MANUSCRIPT AbstractTalkers modify the way they speak in the presence of noise. As well as increases in voice level and fundamental frequency (F0), a flattening of spectral tilt is observed.The resulting "Lombard speech" is typically more intelligible than speech produced in quiet, even when level differences are removed. What is the cause of the enhanced intelligibility of Lombard speech? The current study explored the relative contributions to intelligibility of changes in mean F0 and spectral tilt. The roles of F0and spectral tilt were assessed by measuring the intelligibility gain of non-Lombard speech whose mean F0 and spectrum were manipulated, both independently and in concert, to simulate those of natural Lombard speech. In the presence of speechshaped noise, flattening of spectral tilt contributed greatly to the intelligibility gain of noise-induced speech over speech produced in quiet while an increase in F0 did not have a significant influence. The perceptual effects of spectrum flattening was attributed to its ability of increasing the amount of speech time-frequency plane "glimpsed" in the presence of noise. However, spectral tilt changes alone could not fully account for the intelligibility of Lombard speech. Other changes observed in Lombard speech such as durational modifications may well contribute to intelligibility.

Voice Use Among Music Theory Teachers: A Voice Dosimetry and Self-Assessment Study

“…For both genders, these results showed a strong positive correlation between background noise level and voice SPL, which is in line with past findings 16,30,31 and reflects the Lombard effect. We assume that, for the sake of intelligibility, teachers raise their voice in response to increased background noise.…”

Analysis and HMM-based synthesis of hypo and hyperarticulated speech

“…It is indeed known that a speaker tends to increase his vocal efforts to be more easily understood while talking in a background noise [Summers et al (1988)]. Various aspects of the Lombard effect were already studied, including acoustic and articulatory characteristics [Garnier et al (2006a)] [Garnier et al (2006b)], features extracted from the glottal flow ], or changes of F0 and of the spectral tilt [Lu and Cooke (2009)…”

Evaluation of Spectral Tilt Measures for Sentence Prominence Under Different Noise Conditions

“…In another study, Campbell and Beckman used the harmonic ratio (difference in dB between the first and second harmonic of F0, H1-H2) in order to quantify a measure for spectral tilt [11]. Other studies use an array of different methods, including calculation of the difference in dB between the overall intensity and the intensity of the fundamental frequency (or in a frequency band centered at the fundamental) [12,13,14], taking the first cepstral coefficient [15], taking the difference in dB between a signal with highfrequency pre-emphasis and flat frequency weighting (SPHL-SPL) [16], taking the difference in dB between the first harmonic and third formant (H1-F3) [17], fitting a regression line in the magnitude spectrum [18,19], taking the bandlimited spectral energy ratios [20,21], using the long-term average spectrum (LTAS) to obtain band-limited energy ratios [22], and using all-pole modeling techniques [23]. In addition, some studies utilize measures such as regression line fitting and harmonic ratio, but, instead of applying the measures directly on the short-term spectrum of speech, they utilize the spectrum of the glottal source waveform that is obtained through glottal inverse filtering (GIF) [24,25].…”