The potential role of speech production models in automatic speech recognition

1996 ◽  
Vol 99 (3) ◽  
pp. 1699-1709 ◽  
Author(s):  
R. C. Rose ◽  
J. Schroeter ◽  
M. M. Sondhi
Keyword(s):  
Speech Recognition ◽  
Speech Production ◽  
Automatic Speech Recognition ◽  
Potential Role ◽  
Production Models

scholarly journals Speech production knowledge in automatic speech recognition

2007 ◽  
Vol 121 (2) ◽  
pp. 723-742 ◽  
Author(s):  
Simon King ◽  
Joe Frankel ◽  
Karen Livescu ◽  
Erik McDermott ◽  
Korin Richmond ◽  
...  
Keyword(s):  
Speech Recognition ◽  
Speech Production ◽  
Automatic Speech Recognition ◽  
Production Knowledge

scholarly journals Sounds over Symbols? The Role of Auditory Cues in Orthographically-Correlated Speech Behavior

Languages ◽  
2019 ◽  
Vol 4 (3) ◽  
pp. 72
Author(s):  
Shannon Grippando
Keyword(s):  
Lower Bound ◽  
Speech Production ◽  
Significant Role ◽  
Critical Levels ◽  
Auditory Cues ◽  
Audio Cues ◽  
Speech Patterns ◽  
Production Models ◽  
Written Form

A recent series of studies found a correlation between orthographic length and speech duration: The more orthographic units in a written form, the longer the speech duration of that word, all else being equal. Modular and encapsulated speech production models argue that orthography should not contribute to articulation when it is not directly and explicitly relevant to speech. Such models demand that other factors such as auditory cues must be contributing to the development of this behavior. If auditory cues are being used in the development of these speech patterns, individuals would be expected to be sensitive to these differences. The current study uses an ABX task to determine whether participants are sensitive to durational differences at lengths similar to those observed in the previously found orthographically-correlated speech behavior. The current results showed no sensitivity to the critical levels of speech duration. Participants only began to show sensitivity at four times the length of the lower-bound durational lengths previously observed in individual’s speech patterns. These results call into question whether audio cues are playing a significant role in the development of this speech behavior and strengthen the claim that orthography may be influencing speech in an interactive fashion.

Studying the role of pitch-adaptive spectral estimation and speaking-rate normalization in automatic speech recognition

2018 ◽  
Vol 79 ◽  
pp. 142-151 ◽  
Author(s):  
S. Shahnawazuddin ◽  
Nagaraj Adiga ◽  
Hemant K. Kathania ◽  
Gaydhar Pradhan ◽  
Rohit Sinha
Keyword(s):  
Speech Recognition ◽  
Automatic Speech Recognition ◽  
Spectral Estimation ◽  
Speaking Rate
Sign in / Sign up

Export Citation Format

Share Document