Research

LINEでは、AI事業を戦略事業の一つとして位置付け、事業拡大に向けて積極的に展開しています。AI技術の研究・開発およびAI技術を活用した事業の発展を加速させることを目的として、「データ基盤開発」「データ分析」「機械学習」「AI技術開発」「基礎研究」を担うチームを「Data Science and Engineeringセンター」という1つの組織に集約しています。 これにより各チーム間の連携を強め、「研究 > 開発 > 事業化」のサイクルをスピードアップしています。また事業や担当領域を超えた横断的な大規模データを最大限に活用し、新たなAI関連サービス・新機能を創出するとともに、各種サービスのさらなるユーザ体験向上にも注力しています。

具体的な研究領域は、機械学習を軸に、音声処理、言語処理、画像処理などです。研究テーマについては、日本国内の多数の大学と連携し、共同研究を行う体制を備えています。今後は海外の研究機関との連携も視野に入れています。

また研究・開発に没頭できるよう、オフィスはリラックスできる環境が用意されている他、社内外からAI領域の第一人者を招待するイベントが頻繁に開催されているなど、外部からの刺激は多い環境です。研究意欲が旺盛で刺激的な研究環境に飛び込みたい人は、ぜひお越しください。

Publications


HUMAN-COMPUTER INTERACTION Conference

Continuous and Gradual Style Changes of Graphic Designs with Generative Model

  • M. Ueno and S. Satoh
  • IUI2021, 2021/4

MACHINE LEARNING SECURITY & PRIVACY Conference

P3GM: Private High-Dimensional Data Release via Privacy Preserving Phased Generative Model

  • S. Takagi, T. Takahashi, Y. Cao and M. Yoshikawa
  • ICDE2021, 2021/4

SPEECH PROCESSING Conference

Sound Event Localization and Detection using a Recurrent Convolutional Neural Network and Gated Linear Unit

  • T. Komatsu, M. Togami, T. Takahashi
  • EUSIPCO 2020, 2021/1


SPEECH PROCESSING Conference

Robust Acoustic Scene Classification to Multiple Devices Using Maximum Classifier Discrepancy and Knowledge Distillation

  • S. Takeyama, T. Komatsu, K. Miyazaki, M. Togami, S. Ono
  • EUSIPCO 2020, 2021/1

MACHINE LEARNING Workshop

Disentangling Clustered Representations of Variational Autoencoders for Generating Diverse Samples

  • T. Takahashi, T. Komatsu and K. Yamada
  • LDRC at IJCAI2020, 2021/1

SPEECH PROCESSING Conference

Over-determined Speech Source Separation and Dereverberation

  • M. Togami and R. Scheibler
  • APSIPA2020, 2020/12

SPEECH PROCESSING Conference

INTEGRATION OF SEMI-BLIND SPEECH SOURCE SEPARATION AND VOICE ACTIVITY DETECTION FOR FLEXIBLE SPOKEN DIALOGUE

  • M. Wake, M. Togami, K. Yoshii, and T. Kawahara
  • APSIPA2020, 2020/12

SPEECH PROCESSING Conference

Computer-Resource-Aware Deep Speech Separation with a Run-Time-Specified Number of BLSTM Layers

  • M. Togami, Y. Masuyama, T. Komatsu, K. Yoshii, T. Kawahara
  • APSIPA2020, 2020/12

SPEECH PROCESSING Workshop

Conformer-based sound event detection with semi-supervised learning and data augmentation

  • K. Miyazaki, T. Komatsu, T. Hayashi, S. Watanabe, T. Toda, and K. Takeda
  • DCASE2020, 2020/11

SPEECH PROCESSING Conference

Generalized Minimal Distortion Principle for Blind Source Separation

  • R. Scheibler
  • INTERSPEECH2020, 2020/10

SPEECH PROCESSING Conference

Sparseness-Aware DOA Estimation with Majorization Minimization

  • M. Togami and R. Scheibler
  • INTERSPEECH2020, 2020/10

SPEECH PROCESSING Conference

Mentoring-Reverse Mentoring for Unsupervised Multi-channel Speech Source Separation

  • Y. Nakagome, M. Togami, T. Ogawa and T. Kobayashi
  • INTERSPEECH2020, 2020/10

MACHINE LEARNING SECURITY & PRIVACY Workshop

Locally Private Distributed Reinforcement Learning

  • H. Ono and T. Takahashi
  • FL-ICML at ICML2020, 2020/7

COMPUTER VISION Conference

Neural Implicit Embedding for Point Cloud Analysis

  • K. Fujiwara and T. Hashimoto
  • CVPR2020, 2020/6

SPEECH PROCESSING Conference

UNSUPERVISED TRAINING FOR DEEP SPEECH SOURCE SEPARATION WITH KULLBACK-LEIBLER DIVERGENCE BASED PROBABILISTIC LOSS FUNCTION

  • M. Togami, Y. Masuyama, T. Komatsu, and Y. Nakagome
  • ICASSP2020, 2020/5

SPEECH PROCESSING Conference

DEEP SPEECH EXTRACTION WITH TIME-VARYING SPATIAL FILTERING GUIDED BY DESIRED DIRECTION ATTRACTOR

  • Y. Nakagome, M. Togami, T. Ogawa, and T. Kobayashi
  • ICASSP2020, 2020/5

SPEECH PROCESSING Conference

CONSISTENCY-AWARE MULTI-CHANNEL SPEECH ENHANCEMENT USING DEEP NEURAL NETWORKS

  • Y. Masuyama, M. Togami, and T. Komatsu
  • ICASSP2020, 2020/5



SPEECH PROCESSING Conference

SCENE-DEPENDENT ACOUSTIC EVENT DETECTION WITH SCENE CONDITIONING AND FAKE-SCENE-CONDITIONED LOSS

  • T. Komatsu, K. Imoto, M. Togami
  • ICASSP2020, 2020/5

SPEECH PROCESSING Conference

WEAKLY-SUPERVISED SOUND EVENT DETECTION WITH SELF-ATTENTION

  • K. Miyazaki, T. Komatsu, T. Hayashi, S. Watanabe, T. Toda, K. Takeda
  • ICASSP2020, 2020/5


SPEECH PROCESSING Conference

IMPROVING LPCNET-BASED TEXT-TO-SPEECH WITH LINEAR PREDICTION-STRUCTURED MIXTURE DENSITY NETWORK

  • M.-J. Hwang, E. Song, R. Yamamoto, F. Soong, H.-G. Kang
  • ICASSP2020, 2020/5

SPEECH PROCESSING Conference

ESPNET-TTS: UNIFIED, REPRODUCIBLE, AND INTEGRATABLE OPEN SOURCE END-TO-END TEXT-TO-SPEECH TOOLKIT

  • T. Hayashi, R. Yamamoto, K. Inoue, T. Yoshimura, S. Watanabe, T. Toda, K. Takeda, Y. Zhang, X. Tan
  • ICASSP2020, 2020/5

SPEECH PROCESSING Conference

SEMI-SUPERVISED SPEAKER ADAPTATION FOR END-TO-END SPEECH SYNTHESIS WITH PRETRAINED MODELS

  • K. Inoue, S. Hara, M. Abe, T. Hayashi, R. Yamamoto, S. Watanabe
  • ICASSP2020, 2020/5

SPEECH PROCESSING Conference

Neural Text-to-Speech with a Modeling-by-Generation Excitation Vocoder

  • E. Song, M-J. Hwang, R. Yamamoto, J-S. Kim, O. Kwon, and J-M. Kim
  • INTERSPEECH2020, 2020/5

MACHINE LEARNING SECURITY & PRIVACY Conference

Indirect Adversarial Attacks via Poisoning Neighbors for Graph Convolutional Networks

  • T. Takahashi
  • BigData2019, 2019/12


SPEECH PROCESSING Conference

Multichannel Loss Function for Supervised Speech Source Separation by Mask-based Beamforming

  • Y. Masuyama, M. Togami, T. Komatsu
  • INTERSPEECH2019, 2019/9


NLP Conference

FAQ Retrieval using Query-Question Similarity and BERT-Based Query-Answer Relevance

  • W. Sakata, T. Shibata, R. Tanaka, and S. Kurohashi
  • SIGIR2019, 2019/7


SPEECH PROCESSING Conference

Spatial constraint on multi-channel deep clustering

  • M. Togami
  • ICASSP2019, 2019/5

SPEECH PROCESSING Conference

Multi-channel Itakura Saito Distance Minimization with deep neural network

  • M. Togami
  • ICASSP2019, 2019/5