19. Keio University Japanese Emotional Speech Database (Keio-ESD)

Data DOI

https://doi.org/10.32130/src.Keio-ESD

Producer, Project

Assoc. Prof. Tsuyoshi Moriyama, Tokyo Polytechnic University

Contents

A set of human speech with vocal emotion spoken by a Japanese male speaker and a set of artificial speech that were synthesized by a system that had been developed using the subset of this database for training.

• Examples of recorded words

  # of syllables	accent types	words
  3	HLL	/midori/
  	LHL	/naname/
  	LHH	/nagame/
  4	HLLL	/arawani/
  	LHLL	/amamizu/
  	LHHL	/arayuru/
  	LHHH	/omonaga/

• Emotions
  • Angry, joyful, disgusting, downgrading, funny, worried, gentle, relief, indignation, shameful, etc.（47 emotions）

Speaker

One male speaker

Recording environment

Soundproof room

Speech file format

WAV format (16 kHz, 16 bit, Mono)

Distribution media

1 CD-ROM

Licensing

For research purpose only

Price

No fee

Supplements

Syllable bundaries determined by visual inspection for each speech file are attached.

Further information

http://www.mega.t-kougei.ac.jp/contents-design/moriyama/projects/emotion-synthesis-e.htm

Speech sample for test listening

Word /midori/: 3 morae, accent type "HLL"

angry / joyful / disgusting / surprised / downgrading

Word /amanogawa/: 5 morae, accent type "LHHLL"

disappointing / funny / sad / bored / suffering