Multilingual Speech Processing

498 INDEX

intonation languages, 15, 219

IPA (International Phonetic Alphabet), 12, 92,

162, 278

ISCII (Indian Script Code for Information

Interchange), 28

ISIP (Institute for Signal and Information

Processing), 44, 54

ISIS system, 407–417

discourse and dialog, 412–413

example dialog in, 417

ISIS domain, 407–408

multiagent communication in, 426–427

natural language understanding (NLU),

409–412

overview, 407

speaker authentication (SA), 414

spoken response generation, 413

system architecture—CORBA, 414–417

trilingual speech recognition (SR), 408–409

ISLE Corpus, 286

isolating languages, 17

IST (Information Society Technologies)

program, 57

IWSLT (International Workshop for Spoken

Language Translation), 371

J

JANUS speech recognition tool kit (JRTk), 344

Japanese speech recognition, as ATR system

component, 376–378

JEIDA, 44

join costs, 212

Joint Research Centre (JRC), 56

JRC (Joint Research Centre), 56

JRTk (JANUS speech recognition tool kit), 344

K

Knowledge Query and Manipulation Language

(KQML), 425

knowledge scope expansion, adaptivity to,

417–425

detecting new stock names, 420–422

invoking subdialog for new stock

incorporation, 422–425

overview, 417–420

Korean and Farsi CallFriend calls, 46

KQML (Knowledge Query and Manipulation

Language), 425

Kullback-Leibler divergence, 177

L

LangNet project, 43

language characteristics, 5–31

language in context, 20–22

languages and dialects, 5–8

languages and speech technology, 30–31

linguistic description and classification, 8–20

language family, 9–11

language typology, 11–20

writing systems, 22–29

digital representation and input methods,

27–28

oral languages, 28–29

types of, 23–27

language detection, 242

Language Engineering program, 57

language family, 8, 9–11

language identification (LID), 46; see also

automatic language identification

language independent acoustic modeling,

102–110

combination of context-dependent acoustic

models, 106–110

data-driven model combination, 104–105

heuristic model combination, 104

hierarchical model combination, 105–106

overview, 102–103

language independent sound inventories and

representations, 91–102

alternative representations, 93–96

overview, 91–92

phonetic coverage across languages,

99–102

phonetic sound inventories, 92–93

sound unit sharing across languages, 96–99

language model for trees, relating to statistical

translation using formal Interlingua, 331

language modeling, multilingual, see

multilingual language modeling

language portability, relating to

speech-to-speech translation, 319–320

language resources, 67–70

Language Resources and Evaluation (LRE)

program, 57