ELRA - ELRA-U-S 0097 : Expressive Speech Processing Corpus

You are here » Universal Catalogue » Spoken Resources » Desktop/microphone

Language Resources

Search Catalogue

Send us information

Would you like to collaborate ?
Contact Us

Languages

Catalog Reference : ELRA-U-S 0097

Expressive Speech Processing Corpus

This expressive speech corpus contains longitudinal data of natural conversational speech in Japanese. It gathers the recordings of 50 speakers aged from 3 months to fifty years old, for a total of 1,500 hours of manually transcribed speech (recorded over 5 years).

Recording material: combination of a Sennheiser HMD410 head-mounted dynamic microphone and a SONY Minidisk recorder for most of the recordings.

File characteristics: all files were downsampled to 16kHz at 16-bit.

Transcription and annotation: manual transcription (time-aligned with the audio files) and labeling of information on emotion, interlocutor, dialogue-act, manner (polite, casual, formal, informal), mood and voice (breathy, hard, soft, dark, bright, etc).

The aim was to collect longitudinal data of natural conversational speech to illustrate the wide range of speaking-styles that can occur in ordinary everyday conversational situations.

Production

Project : Expressive Speech Processing Project

Creation date : 2004

Applications

	Applications possible : Speech synthesis
application Area : Research

Contents

Click on the arrow to display content.

speech corpus
Language(s) : Japanese
Duration : 1,500 hours
Quantisation : 16 bits
Source Channel : Microphone
Speech Acquisition Mode : Acoustic
Transcription Entries : Orthographic