Presentation | 2017-03-02 [Poster Presentation] Use of the end of sentence and speaker-derived information in recurrent neural network language models for multiparty conversations. Hiroto Ashikawa, Naohiro Tawara, Atsunori Ogawa, Tomoharu Iwata, Tetsuji Ogawa, Tetsunori Kobayashi, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Information on the end of sentence (EOS) and speaker alternation was exploited in recurrent neural network-based language models and its contribution to improvement in performance of predicting subsequent wordsin multiparty conversations was investigated. These kinds of information were represented as context cues and feature vectors. The former context cues can be inserted to the transcriptions for training, equivalently as other word tokens. The latter feature vectors can be taken as inputs to the neural networks. Experimental comparisons using actual multiparty conversations demonstrated that both representations reduced the perplexity compared tothe case without the EOS and speaker information. The EOS information contributed a lot to improvement in performance of word prediction, especially for multiparty conversations. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | recurrent neural network / context cue / language model / multiparty conversation |
Paper # | EA2016-133,SIP2016-188,SP2016-128 |
Date of Issue | 2017-02-22 (EA, SIP, SP) |
Conference Information | |
Committee | SP / SIP / EA |
---|---|
Conference Date | 2017/3/1(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Okinawa Industry Support Center |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Speech, Engineering/Electro Acoustics, Signal Processing, and Related Topics |
Chair | Kazunori Mano(Shibaura Inst. of Tech.) / Makoto Nakashizuka(Chiba Inst. of Tech.) / Mitsunori Mizumachi(Kyushu Inst. of Tech.) |
Vice Chair | Hiroki Mori(Utsunomiya Univ.) / Masahiro Okuda(Univ. of Kitakyushu) / Shogo Muramatsu(Niigata Univ.) / Yoichi Haneda(Univ. of Electro-Comm.) / Suehiro Shimauchi(NTT) |
Secretary | Hiroki Mori(Kobe Univ.) / Masahiro Okuda(Shizuoka Univ.) / Shogo Muramatsu(Ritsumeikan Univ.) / Yoichi Haneda(Chiba Inst. of Tech.) / Suehiro Shimauchi(KDDI R&D Labs.) |
Assistant | Taichi Asami(NTT) / Kei Hashimoto(Nagoya Inst. of Tech.) / Osamu Watanabe(Takushoku Univ.) / Shigeto Takeoka(Shizuoka Inst. of Science and Tech.) / TREVINO Jorge(Tohoku Univ.) |
Paper Information | |
Registration To | Technical Committee on Speech / Technical Committee on Signal Processing / Technical Committee on Engineering Acoustics |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | [Poster Presentation] Use of the end of sentence and speaker-derived information in recurrent neural network language models for multiparty conversations. |
Sub Title (in English) | |
Keyword(1) | recurrent neural network |
Keyword(2) | context cue |
Keyword(3) | language model |
Keyword(4) | multiparty conversation |
1st Author's Name | Hiroto Ashikawa |
1st Author's Affiliation | Waseda University(Waseda Univ.) |
2nd Author's Name | Naohiro Tawara |
2nd Author's Affiliation | Waseda University(Waseda Univ.) |
3rd Author's Name | Atsunori Ogawa |
3rd Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
4th Author's Name | Tomoharu Iwata |
4th Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
5th Author's Name | Tetsuji Ogawa |
5th Author's Affiliation | Waseda University(Waseda Univ.) |
6th Author's Name | Tetsunori Kobayashi |
6th Author's Affiliation | Waseda University(Waseda Univ.) |
Date | 2017-03-02 |
Paper # | EA2016-133,SIP2016-188,SP2016-128 |
Volume (vol) | vol.116 |
Number (no) | EA-475,SIP-476,SP-477 |
Page | pp.pp.287-290(EA), pp.287-290(SIP), pp.287-290(SP), |
#Pages | 4 |
Date of Issue | 2017-02-22 (EA, SIP, SP) |