Spatial separation of speech signals using amplitude estimation based on interaural comparisons of zero-crossings

2009 ◽  
Vol 51 (1) ◽  
pp. 15-25 ◽  
Author(s):  
Hyung-Min Park ◽  
Richard M. Stern
Keyword(s):  
Spatial Separation ◽  
Speech Signals ◽  
Zero Crossings ◽  
Amplitude Estimation

The Effect of Nonlinear Amplitude Growth on the Speech Perception Benefits Provided by a Single-Sided Vocoder

2019 ◽  
Vol 62 (3) ◽  
pp. 745-757 ◽  
Author(s):  
Jessica M. Wess ◽  
Joshua G. W. Bernstein
Keyword(s):  
Transfer Functions ◽  
Spatial Configuration ◽  
Free Field ◽  
Spatial Separation ◽  
Compression And Expansion ◽  
Interaural Difference ◽  
Interaural Differences ◽  
Single Sided Deafness ◽  
Perceptual Separation ◽  
Loudness Growth

PurposeFor listeners with single-sided deafness, a cochlear implant (CI) can improve speech understanding by giving the listener access to the ear with the better target-to-masker ratio (TMR; head shadow) or by providing interaural difference cues to facilitate the perceptual separation of concurrent talkers (squelch). CI simulations presented to listeners with normal hearing examined how these benefits could be affected by interaural differences in loudness growth in a speech-on-speech masking task.MethodExperiment 1 examined a target–masker spatial configuration where the vocoded ear had a poorer TMR than the nonvocoded ear. Experiment 2 examined the reverse configuration. Generic head-related transfer functions simulated free-field listening. Compression or expansion was applied independently to each vocoder channel (power-law exponents: 0.25, 0.5, 1, 1.5, or 2).ResultsCompression reduced the benefit provided by the vocoder ear in both experiments. There was some evidence that expansion increased squelch in Experiment 1 but reduced the benefit in Experiment 2 where the vocoder ear provided a combination of head-shadow and squelch benefits.ConclusionsThe effects of compression and expansion are interpreted in terms of envelope distortion and changes in the vocoded-ear TMR (for head shadow) or changes in perceived target–masker spatial separation (for squelch). The compression parameter is a candidate for clinical optimization to improve single-sided deafness CI outcomes.

Acoustic Analysis of Emotional Expressions Present in Speech Signals.

1999 ◽  
Vol 53 (5) ◽  
pp. 769-772
Author(s):  
Yoshito Mekada ◽  
Miyuki Mukasa ◽  
Hiroshi Hasegawa ◽  
Masao Kasuga ◽  
Shuichi Matsumoto ◽  
...  
Keyword(s):  
Acoustic Analysis ◽  
Emotional Expressions ◽  
Speech Signals
Sign in / Sign up

Export Citation Format

Share Document