MagicData
SIGN IN

Dataset Overview

Dataset Type

ASR speech corpus

Language

zh-CN

Speech Style

Scripted

Content

keyword spotting, command and query, digits, and news

Audio Parameters

48 kHz, 16 bits, mono

File Format

WAV (PCM)
TXT (UTF8)

Recording Equipment

microphone

Recording Environment

indoor environment
Proprietary
ASRデータセット
477 hours

ASR-BigSCCompreSC: A Scripted Chinese Comprehensive-use Speech Corpus

MDT-ASR-F055 | 477 hours of transcribed Mandarin Chinese scripted speech on keyword spotting, command and query, digits, and news

This dataset consists of 477 hours of transcribed Mandarin Chinese scripted speech focusing on keyword spotting, command and query, digits, and news contributed by 1,586 speakers.

連絡先 business@magicdatatech.com to learn more.

Sample:

Dataset Overview

Dataset Type

ASR speech corpus

Language

zh-CN

Speech Style

Scripted

Content

keyword spotting, command and query, digits, and news

Audio Parameters

48 kHz, 16 bits, mono

File Format

WAV (PCM)
TXT (UTF8)

Recording Equipment

microphone

Recording Environment

indoor environment

License

{{ reviewsTotal }}{{ options.labels.singularReviewCountLabel }}
{{ reviewsTotal }}{{ options.labels.pluralReviewCountLabel }}
{{ options.labels.newReviewButton }}
{{ userData.canReview.message }}

Verifying Email