ELRA - ELRA-U-S 0037 : Mandarin Speech across Taiwan DataBase

You are here » Universal Catalogue » Spoken Resources » Telephone

Language Resources

Search Catalogue

Send us information

Would you like to collaborate ?
Contact Us

Languages

Catalog Reference : ELRA-U-S 0037

Mandarin Speech across Taiwan DataBase

Within the framework of a project sponsored by the National Science Council of Taiwan, a speech database of 7,000 speakers was collected through telephone networks in Taiwan. Nine recording stations were set up on the network and worked during three years (1995-1998). The aim was to collect material from Mandarin speakers in Taiwan, to improve speech generation and recognition. It is divided into the following subsets:

MATDB-1: Short answers statements (spontaneous)
MATDB-2: Numbers pronounced in five ways (read)
MATDB-3: Mandarin syllables (read)
MATDB-4: Words of 2 to 4 syllables (read)
MATDB-5: Phonetically balanced sentences (read)
MATDB-6: Numbers (read)
MATDB-7: Special terms (read)
MATDB-8: Words of 2 to 4 syllables (read)
MATDB-9: Sentences (read)

With the collected material, several databases were produced for different purposes: MAT-800, MAT-160, MAT-2400 (including MAT-400), MAT-2000, MAT-2500.

Production

Project : MAT Project

Applications

	Applications possible : Speech recognition#Speech synthesis
application Area : Research

Contents

Click on the arrow to display content.

speech corpus
Language(s) : Mandarin (Taiwan)
Signal Encoding : Linear PCM
Source Channel : Telephone
Speech Acquisition Interface : Sound card
Speech Acquisition Mode : Acoustic
Recording Period : 1995-1998
Speech Content : Application words#Digit-set#Isolated digits#Isolated words#Natural numbers#Syllables#Phonetically balanced sentences