| Presentation | 2014-06-14 Automatically-Generated Dictionary-based Emotion Recognition from Tweet Speech Eri YANASE, Hiromitsu NISHIZAKI, Yoshihiro SEKIGUCHI, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We make a study of utilization of linguistic information for classification of emotion in spoken utterances. In Japanese, there are a few researches on emotion recognition using linguistic features extracted from a spoken utterance. This paper describes an emotion recognition method that is based on an automatic construction of an emotional words/phrases dictionary from Twitter and an emotion recognition experiment using the dictionary. The dictionary is made by collecting so many tweets from Twitter, and the extracted words/phrases are registered to the dictionary. For an emotion recognition experiment, we prepared two types of test data; text tweets and spoken tweets. For the spoken tweets, we try two types of emotional words/phrases extraction methods; one is to use an automatic speech recognizer and the other is to use a spoken term detection technique. The experimental result showed that the emotion recognition accuracy was about 40% for the text tweets, and the accuracy for the spoken tweets was slightly lower than the one of the text tweets. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Emotion recognition / linguistic information / speech recognition / spoken term detection |
| Paper # | NLC2014-4 |
| Date of Issue |
| Conference Information | |
| Committee | NLC |
|---|---|
| Conference Date | 2014/6/7(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Natural Language Understanding and Models of Communication (NLC) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Automatically-Generated Dictionary-based Emotion Recognition from Tweet Speech |
| Sub Title (in English) | |
| Keyword(1) | Emotion recognition |
| Keyword(2) | linguistic information |
| Keyword(3) | speech recognition |
| Keyword(4) | spoken term detection |
| 1st Author's Name | Eri YANASE |
| 1st Author's Affiliation | Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi() |
| 2nd Author's Name | Hiromitsu NISHIZAKI |
| 2nd Author's Affiliation | Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi |
| 3rd Author's Name | Yoshihiro SEKIGUCHI |
| 3rd Author's Affiliation | Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi |
| Date | 2014-06-14 |
| Paper # | NLC2014-4 |
| Volume (vol) | vol.114 |
| Number (no) | 81 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |