Papers by Yuichi Ishimoto

4 papers
Extending Search System based on Interactive Visualization for Speech Corpora (L18-1)

Copied to clipboard

Challenge: Speech corpora are indispensable to speech research; data centers are set up to meet this demand . it is difficult for speech corpus users to compare and select suitable corporata from the large variety of languages .
Approach: They propose a search system that allows users to search for speech corpora interactively and visually . they add specification attributes and items to a large-scale metadata database "SHACHI"
Outcome: The proposed system can search speech corpora interactively and visually . it would be easier for users to select suitable corporata from the large number of languages available .
KOTONOHA: A Corpus Concordance System for Skewer-Searching NINJAL Corpora (2020.lrec-1)

Copied to clipboard

Challenge: NINJAL has developed several types of corpora for linguistic research . for each corpus NINJAL provided an online search environment, ‘Chunagon’ .
Approach: NINJAL has developed several types of corpora for linguistic research . for each corpus NINJAL provided an online search environment, ‘Chunagon’, which is a morphological-information-annotation-based concordance system made publicly available in 2011 . NINjal has now provided a system ‘Kotonoha’ based on the ‘Chunegon’ systems .
Outcome: NINJAL has provided a skewer-search system ‘Kotonoha’ based on ‘Chunagon’ systems.
A Conversation-Analytic Annotation of Turn-Taking Behavior in Japanese Multi-Party Conversation and its Preliminary Analysis (2020.lrec-1)

Copied to clipboard

Challenge: a new conversation-analytic annotation scheme is proposed for multi-party conversations . current systems do not take a turn like a human even in simple two-party conversation .
Approach: They propose a conversation-analytic annotation scheme for turn-taking behavior in multi-party conversations . they analyze how syntactic and prosodic features of utterances vary across four selection types .
Outcome: The proposed model is based on Japanese multi-party conversations.
Design and Evaluation of the Corpus of Everyday Japanese Conversation (2022.lrec-1)

Copied to clipboard

Challenge: a corpus of everyday conversations that includes video data is a new approach . a large corpus contains 200 hours of speech, 577 conversations, about 2.4 million words .
Approach: They have constructed a corpus of everyday conversations that includes video data . they will publish the corpus in march 2022, and part of it in 2018 on trial basis .
Outcome: The corpus of everyday Japanese conversation (CEJC) contains audio and video data . the study shows that the corpus includes a good balance of adult conversants .

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations