研究開発

 リサーチチームは、世界最先端の研究開発によりAIを通じた未来の変革を行うrinna株式会社を支えます。 主にテキスト・音声・画像に関する研究を行っており、トップカンファレンスなどにも採択されています。

発表論文やビジネスに関するお問合せは、論文の著者ではなくこちらのフォームからお問合せください。
For inquiries regarding published papers and business, please use this form instead of the paper authors.

発表論文一覧

-2022-

• Kentaro Mitsui, Tianyu Zhao, Kei Sawada, Yukiya Hono, Yoshihiko Nankaku, Keiichi Tokuda, “End-to-End Text-to-Speech Based on Latent Representation of Speaking Styles Using Spontaneous Dialogue”, INTERSPEECH 2022, September 2022. [arXiv] [Demo]
• Kentaro Mitsui, Kei Sawada, “MSR-NV: Neural Vocoder Using Multiple Sampling Rates”, INTERSPEECH 2022, September 2022. [arXiv] [Demo]
• 三井健太郎, 趙天雨, 沢田慶, 法野行哉, 南角吉彦, 徳田恵一, “自発的対話を用いた潜在スタイル表現の抽出・予測に基づく音声合成”, 日本音響学会2022年秋季研究発表会, 2022年9月.
• 三井健太郎, 沢田慶, “テキストを入力とする音声・顔ランドマーク系列の同期生成”, 日本音響学会2022年秋季研究発表会, 2022年9月.
• 沢田慶, シーン誠, 三井健太郎, 趙天雨, “ディープラーニングの活用：AI × キャラクターによる新しいゲームの世界”, コンピュータエンターテインメントデベロッパーズカンファレンス2022 (CEDEC2022), 2022年8月.
• シーン誠, 趙天雨, 沢田慶, “日本語における言語画像事前学習モデルの構築と公開”, 第25回 画像の認識・理解シンポジウム (MIRU2022), 2022年7月. [GitHub] [Hugging Face]
• 三井健太郎, 沢田慶, “MSR-NV: 複数サンプリングレートを用いたニューラルボコーダの検討”, 日本音響学会2022年春季研究発表会, pp. 931-934, 2022年3月.

-2021-

• 趙天雨, 沢田慶, “日本語自然言語処理における事前学習モデルの公開”,  第93回 言語・音声理解と対話処理研究会, pp. 169-170, 2021年11月. [Paper] [GitHub] [Hugging Face]
• 沢田慶, “身近になった対話システム：4．一般ユーザとの雑談会話のためのAIチャットボット”, 情報処理, pp. e19-e23, 2021年9月. [Paper]
• Tianyu Zhao,Tatsuya Kawahara, “Multi-Referenced Training for Dialogue Response Generation”, The 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL 2021), pp. 190-201, July 2021. [Paper] [GitHub] [arXiv]
• Ruozi Huang, Huang Hu, Wei Wu, Kei Sawada, Mi Zhang, Daxin Jiang, “Dance Revolution: Long-Term Dance Generation with Music via Curriculum Learning”, The Ninth International Conference on Learning Representations (ICLR 2021), May 2021. [Paper] [GitHub] [YouTube] [arXiv]
• Ze Yang, Wei Wu, Huang Hu, Can Xu, Wei Wang, Zhoujun Li, “Open Domain Dialogue Generation with Latent Images”, The 35th AAAI Conference on Artificial Intelligence (AAAI-21), pp. 14239-14247, February 2021. [Paper] [arXiv]

- 2020 -

• Linxiao Li, Can Xu, Wei Wu, Yufan Zhao, Xueliang Zhao, Chongyang Tao, “Zero-Resource Knowledge-Grounded Dialogue Generation”, The Thirty-fourth Annual Conference on Neural Information Processing Systems (NeurIPS 2020), December 2020. [Paper] [GitHub] [arXiv]
• Yufan Zhao, Can Xu, Wei Wu, Lei Yu, “Learning a Simple and Effective Model for Multi-turn Response Generation with Auxiliary Tasks”, The 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP 2020), pp. 3472-3483, November 2020. [Paper] [arXiv]
• Xueliang Zhao, Wei Wu, Can Xu, Chongyang Tao, Dongyan Zhao, Rui Yan, “Knowledge-Grounded Dialogue Generation with Pre-trained Language Models”, The 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP 2020), pp. 3377-3390, November 2020. [Paper] [arXiv]
• Ze Yang, Wei Wu, Can Xu, Xinnian Liang, Jiaqi Bai, Liran Wang, Wei Wang, Zhoujun Li, “StyleDGPT: Stylized Response Generation with Pre-trained Language Models”, Findings of the Association for Computational Linguistics: EMNLP 2020, pp. 1548-1559, November 2020. [Paper] [arXiv]
• Yukiya Hono, Kazuna Tsuboi, Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, Keiichi Tokuda, “Hierarchical Multi-Grained Generative Model for Expressive Speech Synthesis”, pp. 3441-3445, INTERSPEECH 2020, October 2020. [Paper] [Demo] [arXiv]
• 法野行哉, 坪井一菜, 沢田慶, 橋本佳, 大浦圭一郎, 南角吉彦, 徳田恵一, “階層化多重粒度生成モデルを用いた表現豊かな音声合成”, 日本音響学会2020年秋季研究発表会, pp. 791-794, 2020年9月.
• Chongyang Tao, Wei Wu, Can Xu, Yansong Feng, Dongyan Zhao, Rui Yan, “Improving Matching Models with Hierarchical Contextualized Representations for Multi-turn Response Selection”, 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2020), pp. 1865-1868, July 2020. [Paper] [arXiv]
• Xueliang Zhao, Wei Wu, Chongyang Tao, Can Xu, Dongyan Zhao, Rui Yan, “Low-Resource Knowledge-Grounded Dialogue Generation”, The Eighth International Conference on Learning Representations (ICLR 2020), May 2020. [Paper] [arXiv]
• 三井健太郎, 法野行哉, 坪井一菜, 沢田慶, “カスケード構造を用いた音声パラメータ予測に基づく統計的パラメトリック音声合成”, 日本音響学会2020年春季研究発表会, pp. 1107-1108, 2020年3月.

- 2019 -

• Ze Yang, Can Xu, Wei Wu, Zhoujun Li, “Read, Attend and Comment: A Deep Architecture for Automatic News Comment Generation”, 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP 2019), pp. 5077-5089, November 2019. [Paper] [arXiv]
• Ze Yang, Wei Wu, Jian Yang, Can Xu, Zhoujun Li, “Low-Resource Response Generation with Template Prior”, 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP 2019), pp. 1886-1897, November 2019. [Paper] [GitHub] [arXiv]
• Jia Li, Chongyang Tao, Wei Wu, Yansong Feng, Dongyan Zhao, Rui Yan, “Sampling Matters! An Empirical Study of Negative Sampling Strategies for Learning of Matching Models in Retrieval-based Dialogue Systems”, 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP 2019), pp. 1291-1296, November 2019. [Paper]
• 坪井一菜, 沢田慶, AIりんな, “AI「りんな」のボイストレーニング”, コンピュータエンターテインメントデベロッパーズカンファレンス2019 (CEDEC2019), 2019年9月. [Slide]
• Xueliang Zhao, Chongyang Tao, Wei Wu, Can Xu, Dongyan Zhao, Rui Yan, “A Document-grounded Matching Network for Response Selection in Retrieval-based Chatbots”, The Twenty-Eighth International Joint Conference on Artificial Intelligence (IJCAI 2019), pp. 5443-5449, August 2019. [Paper] [arXiv]
• Can Xu, Wei Wu, Chongyang Tao, Huang Hu, Matt Schuerman, Ying Wang, “Neural Response Generation with Meta-words”, The 57th Annual Meeting of the Association for Computational Linguistics (ACL 2019), pp. 5416–5426, July 2019. [Paper] [arXiv]
• Jiazhan Feng, Chongyang Tao, Wei Wu, Yansong Feng, Dongyan Zhao, Rui Yan, “Learning a Matching Model with Co-teaching for Multi-turn Response Selection in Retrieval-based Dialogue Systems”, The 57th Annual Meeting of the Association for Computational Linguistics (ACL 2019), pp. 3805-3815, July 2019. [Paper] [arXiv]
• Chongyang Tao, Wei Wu, Can Xu, Wenpeng Hu, Dongyan Zhao, Rui Yan, “One Time of Interaction May Not Be Enough: Go Deep with an Interaction-over-Interaction Network for Response Selection in Dialogues”, The 57th Annual Meeting of the Association for Computational Linguistics (ACL 2019), pp. 1-11, July 2019. [Paper]
• 沢田慶, 坪井一菜, Xianchao Wu, Zhan Chen, 法野行哉, 橋本佳, 大浦圭一郎, 南角吉彦, 徳田恵一, “AI歌手りんな：ユーザ歌唱や楽譜を入力とする歌声合成システム”, 日本音響学会2019年春季研究発表会, pp. 1041-1044, 2019年3月.
• Yu Wu, Wei Wu, Chen Xing, Can Xu, Zhoujun Li, Ming Zhou, “A Sequential Matching Framework for Multi-Turn Response Selection in Retrieval-Based Chatbots”, Computational Linguistics, Volume 45, Issue 1, pp. 163-197, March 2019. [Paper] [arXiv]
• 高木信二, 安藤厚志, 越智景子, 沢田慶, 塩田さやか, 鈴木雅之, 玉森聡, 俵直弘, 福田隆, 増村亮, “国際会議Interspeech2018報告”, 第126回音声言語情報処理研究発表会 (SIG-SLP), 2019年2月. [Paper]
• Chongyang Tao, Wei Wu, Can Xu, Wenpeng Hu, Dongyan Zhao, Rui Yan, “Multi-Representation Fusion Network for Multi-Turn Response Selection in Retrieval-Based Chatbots”, The Twelfth ACM International Conference on Web Search and Data Mining (WSDM ’19), pp.267-275, January 2019. [Paper] [GitHub]

- 2018 -

• Yu Wu, Wei Wu, Zhoujun Li, Ming Zhou, “Response selection with topic clues for retrieval-based chatbots”, Neurocomputing, Volume 316, pp. 251-261, November 2018. [Paper] [arXiv]
• Huang Hu, Xianchao Wu, Bingfeng Luo, Chongyang Tao, Can Xu, Wei Wu, Zhan Chen, “Playing 20 Question Game with Policy-Based Reinforcement Learning”, 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP 2018), pp.3233-3242, October 2018. [Paper] [GitHub] [arXiv]
• Kei Sawada, Takenori Yoshimura, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, Keiichi Tokuda, “The NITech text-to-speech system for the Blizzard Challenge 2018”, Blizzard Challenge 2018 Workshop, September 2018. [Paper] [Demo]
• Yu Wu, Wei Wu, Zhoujun Li, Ming Zhou, “Learning Matching Models with Weak Supervision for Response Selection in Retrieval-based Chatbots”, The 56th Annual Meeting of the Association for Computational Linguistics (ACL 2018), pp. 420-425, July 2018. [Paper] [arXiv]
• Chongyang Tao, Shen Gao, Mingyue Shang, Wei Wu, Dongyan Zhao, Rui Yan, “Get The Point of My Utterance! Learning Towards Effective Responses with Multi-Head Attention Mechanism”, The 27th International Joint Conference on Artificial Intelligence and the 23rd European Conference on Artificial Intelligence (IJCAI-ECAI-18), pp. 4418-4424, July 2018. [Paper]
• Can Xu, Wei Wu, Yu Wu, “Towards Explainable and Controllable Open Domain Dialogue Generation with Dialogue Acts”, arXiv preprint arXiv:1807.07255, July 2018. [arXiv]
• Xianchao Wu, Ander Martinez, Momo Klyen, “Dialog Generation Using Multi-Turn Reasoning Neural Networks”, The 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2018), pp. 2049-2059, June 2018. [Paper]
• Xianchao Wu, Huang Hu, Momo Klyen, Kyohei Tomita, Zhan Chen, “Q20: Rinna Riddles Your Mind by Asking 20 Questions”, 言語処理学会第24回年次大会 (NLP2018), pp. 1312-1315, 2018年3月. [Paper]
• Xianchao Wu, Huang Hu, “Evaluating Rinna’s Mind-reading Feature by Self-playing”, 言語処理学会第24回年次大会 (NLP2018), pp. 1235-1238, 2018年3月. [Paper]
• Chen Xing, Wei Wu, Yu Wu, Ming Zhou, Yalou Huang, Wei-Ying Ma, “Hierarchical Recurrent Attention Network for Response Generation”, The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18), pp. 5610-5617, February 2018. [Paper] [arXiv]
• Yu Wu, Wei Wu, Dejian Yang, Can Xu, Zhoujun Li, Ming Zhou, “Neural Response Generation with Dynamic Vocabularies”, The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18), pp. 5594-5601, February 2018. [Paper] [arXiv]
• Yu Wu, Wei Wu, Zhoujun Li, Ming Zhou, “Knowledge Enhanced Hybrid Neural Network for Text Matching”, The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18), pp. 5586-5593, February 2018. [Paper] [arXiv]

- 2017 -

• Xianchao Wu, Hang Tong, Momo Klyen, “Fine-grained sentiment analysis with 32 dimensions”, The 21st International Conference on Asian Language Processing (IALP 2017), December 2017. [Paper]
• 呉先超, 藤原敬三, 飯田勝也, 冨田恭平, 中島りか, “りんなのキャラボックス: 雑談から商品推薦まで”, 第81回言語・音声理解と対話処理研究会, pp. 62-65, 2017年10月. [Paper]
• Yu Wu, Wei Wu, Chen Xing, Ming Zhou, Zhoujun Li, “Sequential Matching Network: A New Architecture for Multi-turn Response Selection in Retrieval-Based Chatbots”, The 55th Annual Meeting of the Association for Computational Linguistics (ACL 2017), pp. 496-505, July 2017. [Paper] [arXiv]
• Xianchao Wu, Momo Klyen, Kazushige Ito, Zhan Chen, “Haiku Generation Using Deep Neural Networks”, 言語処理学会第23回年次大会 (NLP2017), pp. 1133-1136, 2017年3月. [Paper]
• Xianchao Wu, Yuichiro Kikura, Momo Klyen, Zhan Chen, “Sentiment Analysis with Eight Dimensions for Emotional Chatbots”, 言語処理学会第23回年次大会 (NLP2017), pp. 791-794, 2017年3月. [Paper]

- 2016 -

• 呉先超, 伊藤和重, 飯田勝也, 坪井一菜, クライアン桃, “りんな：女子高生人工知能”, 言語処理学会第22回年次大会 (NLP2016), pp. 306-309, 2016年3月. [Paper]