| Project/Area Number |
09480064
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Toyohashi University of Technology |
|---|
Principal Investigator |
NAKAGAWA Seiichi Toyohashi University of Technology, Faculty of Engineering, Professor, 工学部, 教授 (20115893)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
KAI Atsuhiko Shizuoka University, Faculty of Engineering, Assitant Professor, 工学部, 講師 (60283496)
MINEMATSU Nobuaki Toyohashi University of Technology, Faculty of Engineering, Research Assistant, 工学部, 助手 (90273333)
MASUYAMA Sigeru Toyohashi University of Technology, Faculty of Engineering, Professor, 工学部, 教授 (60173762)
ANDO Akio NHK Laboratory, Sub-Head of Human-Interface Department, 放送技術研究所, 副部長
|
|---|
| Project Period (FY) |
1997 – 1999
|
| Project Status |
Completed (Fiscal Year 1999)
|
| Budget Amount *help |
¥13,100,000 (Direct Cost: ¥13,100,000)
Fiscal Year 1999: ¥4,800,000 (Direct Cost: ¥4,800,000)
Fiscal Year 1998: ¥3,100,000 (Direct Cost: ¥3,100,000)
Fiscal Year 1997: ¥5,200,000 (Direct Cost: ¥5,200,000)
|
|---|
| Keywords | speech recognition / acoustic model / closed caption / dictation / language model / summarization / broadcast news / 大語彙連続音声認識 / ニュース文 / 要約 |
|---|
| Research Abstract |
It is well-known that HMMs only of the basic structure can not capture the correlation among successive frames adequately. In our previous work, to solve this problem, segmental unit HMMs were introduced and their effectiveness was shown. And the integration of Δ cepstrum and ΔΔ cepstrum into the segmental unit HMMs was also found to improve the recognition performance in the work. Firstly, we compared frame-based models and segment-based models. Results showed the effectiveness of the use of segmental features as input vectors. Secondly, we compared syllable-based HMMs and triphone-based HMMs. Recognition experiments showed that syllable-based HMMs are suitable for Japanese.
Next, we developed a method that constructs language models using a task adaptation strategy and idiomatic expressions of news articles. First, we investigated the effect of a task adaptation method of N-gram language model using a limited amount of target articles. Second, we investigated the effect of the language model adaptation method using the latest articles. Third, we investigated the effect of the use of idiomatic expressions as morpheme units, since some specific expressions and idiomatic expressions are frequently observed in news articles. We showed that our proposed three methods were effective for constructing N-gram language models.
Finally, we proposed and evaluated a method for summarizing each sentence in TV news texts written in Japanese. It is not appropriate to select important sentences for abstracting news text, because a news text consists of only a few and long sentences. Then, we tried to reduce redundant parts, which consisted of modifier etc., of each sentence. We used a simple parsing method specialized for news texts so that the syntactical structure was not destroyed. We evaluated this summarizing method by obtaining information by means of questionnaires to 32 examinees.
|
