Dataset Overview

Dataset Type

ASR speech corpus

Language

zh-CN

Speech Style

conversational

Content

Customer-Service Conversation (Finance)

Audio Parameters

8 kHz, 16 bits, mono

File Format

WAV (PCM)
TXT (UTF8)

Recording Equipment

telephony

Recording Environment

indoor environment

Popular Datasets

Proprietary
1602 hours
ASR Corpus
Proprietary
427 hours
ASR Corpus
Proprietary
1386 hours
ASR Corpus
Proprietary
313 hours
ASR Corpus
Open Source
NLP Corpus
2 KB
Proprietary
65 hours
ASR Corpus
Proprietary
973 hours
ASR Corpus
Proprietary
ASR Corpus
459 hours

Mandarin Chinese Conversational Speech Corpus – Customer-Service

MDT-ASR-F053 | 459 hours of transcribed Mandarin Chinese scripted speech on financial customer-service conversation.

Dataset Overview

Dataset Type

ASR speech corpus

Language

zh-CN

Speech Style

conversational

Content

Customer-Service Conversation (Finance)
8 kHz, 16 bits, mono

File Format

WAV (PCM)
TXT (UTF8)

Recording Equipment

telephony

Recording Environment

indoor environment

License

This dataset consists of 459 hours of transcribed Mandarin Chinese scripted speech focusing on customer-service conversation contributed by 2,156 speakers.

Contact business@magicdatatech.com to learn more.

Comments

{{ reviewsTotal }} Review
{{ reviewsTotal }} Reviews
{{ options.labels.newReviewButton }}
{{ userData.canReview.message }}