Abstract | ||
---|---|---|
This paper describes our latest efforts in building a speech rec- ognizer for operating a navigation system through speech instead of typed input. Compared to conventional speech recognition for navigation systems, where the input is usually restricted t o a fixed set of keywords and keyword phrases, complete spontaneous sen- tences are allowed as speech input. We will present the interac- tion of speech input, parsing and the necessary reactions to the re- quested queries. Our system has been trained on German sponta- neous speech data and has been adapted to navigation queries us- ing MLLR. As the system is not restricted to command word input, a parser is necessary to further process the recognized utte rance. We show that within a lab environment our system is able to han- dle arbitrary spontaneous sentences as input to a navigatio n system successfully. The performance of the recognizer measured in word error rate gives a result of 18%. The parser has also been eval uated and yields an error rate of 20%. |
Year | Venue | Keywords |
---|---|---|
1998 | ICSLP | and destinations,hesitations,ungrammatical sen- tences etc.. keywords,2. the system has to be able to handle effects inherent in spon- taneous speech false starts,speech recognition,word error rate,error rate |
Field | DocType | Citations |
Computer science,Voice activity detection,Word error rate,Navigation system,Utterance,Speech recognition,Artificial intelligence,Natural language processing,Parsing,German | Conference | 6 |
PageRank | References | Authors |
1.01 | 2 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Petra Geutner | 1 | 200 | 41.39 |
Matthias Denecke | 2 | 177 | 24.32 |
Uwe Meier | 3 | 138 | 20.81 |
Martin Westphal | 4 | 85 | 14.92 |
Alex Waibel | 5 | 6343 | 1980.68 |