Computer Speech Technology

by Rodman, Robert D.

• ISBN13: 9780890062975
• ISBN10: 0890062978
• Format: Hardcover
• Copyright: 1999-02-01
• Publisher: Artech House

Summary

This new book is the first to provide a truly understandable, non-technical overview of all the major areas in the computer processing of human speech -- speech recognition, speech synthesis, speaker recognition, language identification, lip synchronization, and co-channel separation. It takes a unique, nonmathematical approach in exploring the nature of human language and its impact on the science and methodologies of computer voice technology. In one, easy-to-read source, you gain a deeper understanding of the fundamentals, uses, and applications of the technology itself and of the strengths and weaknesses of various systems. A time-saving glossary of technical terms is included.

Table of Contents

Preface	xv
About Speech	1	(50)
Introduction	1	(1)
How Speech Is Produced	2	(32)
The Vocal Tract	2	(3)
Articulatory Phonetics	5	(9)
Phonetic Alphabets	14	(11)
Prosody and Suprasegmentals	25	(3)
Syllables	28	(1)
Dialects	29	(2)
Languages (Other Than English)	31	(3)
Acoustic Phonetics	34	(8)
Phonemics	42	(5)
Articulatory Processes	47	(4)
References	49	(2)
Representing Speech in the Computer	51	(48)
Introduction	51	(1)
Microphones	52	(1)
Sampling	53	(16)
Sampling Rate	54	(8)
Quantization	62	(7)
Speech Digitization	69	(18)
Wave Form Coders	73	(10)
Voice Coders (Vocoders)	83	(4)
The Frequency Domain	87	(12)
The Game of Jumble: Spectrum-Cepstrum, Frequency-Quefrency, Filtering-Liftering	91	(2)
Spectrograms: A Hybrid Representation of Speech	93	(5)
References	98	(1)
Speech Recognition	99	(78)
Introduction	99	(1)
Speech Recognition: What It Is; What It Isn't	100	(1)
Why Is Speech Recognition Easy for us and Difficult for Our Computers?	101	(2)
A Brief History of Speech Recognition	103	(8)
The Era of ARPA	105	(4)
After ARPA	109	(2)
Three Dimensions of Speech Recognition	111	(6)
Continuous Versus Noncontinuous	111	(2)
Speaker-Independent Versus Speaker-Dependent	113	(1)
Vocabulary Size	114	(1)
Tradeoffs and Interactions	115	(2)
Units of Speech Recognition	117	(4)
Words and Phrases	118	(1)
Syllables	118	(1)
Phonemes	119	(1)
Diphones and Triphones	120	(1)
Representing the Units	121	(5)
Acoustic Features	121	(5)
Comparing the Units	126	(7)
Dynamic Time Warping (DTW)	126	(3)
Hidden Markov Models (HMMs)	129	(4)
Future Challenges I	133	(4)
Artificial Neural Networks (ANNs)	133	(4)
Errors	137	(7)
Types of Errors	138	(4)
Error Tolerances	142	(2)
Performance Evaluation of Speech Recognizers	144	(10)
Error Rates	145	(7)
Other Factors	152	(2)
Error Reduction	154	(6)
Environmental Effects	155	(1)
Human Factors	155	(3)
Subsetting	158	(1)
Vocabulary Selection	159	(1)
Error Detection and Correction	160	(15)
Feedback Systems	161	(3)
Higher Levels of Linguistic Knowledge	164	(6)
Automatic Error Correction	170	(5)
Future Challenges II	175	(2)
References	176	(1)
Speech Synthesis	177	(38)
Introduction and History	177	(2)
Parametric Coding (Electronic Synthesis)	179	(5)
Parameters of Parametric Speech Synthesis	179	(3)
Input Units of Parametric Speech Synthesis	182	(2)
Concatenative Synthesis	184	(6)
Allophone Concatenation	186	(1)
Diphone Concatenation	186	(1)
Demisyllable Concatenation	187	(1)
Waveform of Concatenative Units	188	(2)
Text-to-Speech Processing	190	(16)
Rules and Exceptions	191	(2)
Morphological Analysis	193	(1)
Articulation Effects	194	(2)
Prosody	196	(4)
Special Problems	200	(6)
Concept-to-Speech	206	(2)
Languages of the World	208	(1)
Dialects	209	(1)
Performance Evaluation	209	(4)
Intelligibility	210	(1)
Comprehensibility	211	(1)
Pleasantness/Naturalness	212	(1)
Future Challenges	213	(2)
References	213	(2)
Speaker Recognition, Language Identification, and Lip Synchronization	215	(36)
Speaker Recognition	215	(19)
Speaker Recognition Versus Speech Recognition	216	(1)
Types of Speaker Recognition	217	(1)
Text-Dependent, Text-Independent, and Text-Prompted Speaker Recognition	218	(2)
``Voiceprints''	220	(1)
Methods of Speaker Recognition	221	(8)
Noise	229	(2)
Performance Evaluation of Speaker Recognition Systems	231	(3)
Co-channel Speaker Separation	234	(2)
Language Identification	236	(7)
Four Computational Approaches to Language Identification	238	(3)
Performance Evaluation of Language Identification Systems	241	(2)
Lip Synchronization	243	(6)
Visemes	244	(2)
Mapping Directly From the Speech Signal to Mouth Shapes	246	(3)
Future Challenges	249	(2)
References	249	(2)
Applications in Speech Recognition	251	(26)
Criteria for a Viable Speech Recognition Application	252	(3)
Hands Busy, Eyes Busy	252	(1)
Remoteness	253	(2)
Miniaturization	255	(1)
2001 Won't Be 2001	255	(1)
The Role of Human Factors in Speech Recognition Applications	256	(2)
Application Areas	258	(19)
Assistive Technology	259	(3)
Telecommunications	262	(2)
Command and Control	264	(4)
Data Entry and Retrieval	268	(4)
Education	272	(3)
References	275	(2)
Applications in Speech Synthesis	277	(20)
``At the Tone, the Time Will Be...''	278	(1)
When to use Text-to-Speech; When to use Digitally Recorded Speech	278	(2)
Interactive Voice Response Systems (IVRs)	280	(1)
Human Factors Revisited	281	(1)
Application Areas	282	(15)
Aid for Persons with Disabilities	283	(3)
Education	286	(1)
Emergency Scenarios	287	(1)
En Masse Advisories	288	(1)
Information Retrieval	289	(1)
Information Reporting	290	(2)
Electronic Mail and Fax Readers	292	(2)
In the Dark	294	(1)
Toys and Games	295	(1)
Transportation	295	(1)
Government Services	296	(1)
Disguise	296	(1)
References	296	(1)
Applications in Speaker Recognition, Language Identification, and Lip Synchronization	297	(14)
Applications in Speaker Recognition	298	(9)
Access	298	(3)
Authentication	301	(1)
Monitoring	301	(2)
Fraud Prevention	303	(1)
Forensics	304	(2)
Personal Services	306	(1)
Applications in Language Identification	307	(1)
Telecommunications	307	(1)
Communications Monitoring	307	(1)
Public Information Systems	308	(1)
Applications in Automatic Lip Synching	308	(3)
Animation	308	(2)
References	310	(1)
Glossary	311	(20)
About the Author	331	(2)
Index	333