| Presentation | 2001/12/13 The Method of Extracting Peripheral Features that Achieve High Performance in both Tasks of LVCSR and ISWR Takashi FUKUDA, Tsuneo NITTA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this paper, we propose a feature extractor that improves the performance of isolated-word and continuous speech recognition with CMN. Firstly, local features (LF) and peripheral features (PF) extracted from time-spectrum (TS) patterns and their roles in speech recognition are described. Then, the proposed feature extractors are implemented into a standard HMM-based speech recognition system with modified CMN (MCMN) in which CMN is controlled by a normalized variance of an utterance. Experiments were investigated both in an isolated spoken-word recognition (ISWR) system and a large vocabulary continuous speech recognition (LVCSR) system. Experimental results show that the feature set of MFCC with MCMN and novel PF outperforms the baseline feature set in an LVCSR task, and achieves significant improvement in an ISWR task. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | speech recognition / feature extraction / CMN / peripheral feature / local feature / HMM |
| Paper # | NLC2001-50,SP2001-85 |
| Date of Issue |
| Conference Information | |
| Committee | NLC |
|---|---|
| Conference Date | 2001/12/13(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Natural Language Understanding and Models of Communication (NLC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | The Method of Extracting Peripheral Features that Achieve High Performance in both Tasks of LVCSR and ISWR |
| Sub Title (in English) | |
| Keyword(1) | speech recognition |
| Keyword(2) | feature extraction |
| Keyword(3) | CMN |
| Keyword(4) | peripheral feature |
| Keyword(5) | local feature |
| Keyword(6) | HMM |
| 1st Author's Name | Takashi FUKUDA |
| 1st Author's Affiliation | Graduate School of Engineering, Toyohashi University of Technology() |
| 2nd Author's Name | Tsuneo NITTA |
| 2nd Author's Affiliation | Graduate School of Engineering, Toyohashi University of Technology |
| Date | 2001/12/13 |
| Paper # | NLC2001-50,SP2001-85 |
| Volume (vol) | vol.101 |
| Number (no) | 520 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |