Open Source MagicData-RAMC: A Rich Annotated Mandarin Conversational(RAMC) Speech Dataset (2022-03-31T00:00:00.000000Z)

TL;DR

The MagicData-RAMC corpus contains 180 hours of conversational speech data recorded from native speakers of Mandarin Chinese over mobile phones with a sampling rate of 16 kHz and allows extensive research on a series of speech-related tasks, including automatic speech recognition, speaker diarization, topic detection, keyword search, text-to-speech, etc.

Abstract

This paper introduces a high-quality rich annotated Mandarin conversational (RAMC) speech dataset called MagicData-RAMC. The MagicData-RAMC corpus contains 180 hours of conversational speech data recorded from native speakers of Mandarin Chinese over mobile phones with a sampling rate of 16 kHz. The dialogs in MagicData-RAMC are classified into 15 diversified domains and tagged with topic labels, ranging from science and technology to ordinary life. Accurate transcription and precise speaker voice activity timestamps are manually labeled for each sample. Speakers' detailed information is also provided. As a Mandarin speech dataset designed for dialog scenarios with high quality and rich annotations, MagicData-RAMC enriches the data diversity in the Mandarin speech community and allows extensive research on a series of speech-related tasks, including automatic speech recognition, speaker diarization, topic detection, keyword search, text-to-speech, etc. We also conduct several relevant tasks and provide experimental results to help evaluate the dataset.

Authors

Pengyuan Zhang

3 papers

Lei Xie

9 papers

Zehui Yang

1 papers

TL;DR

Abstract

Authors

References37 items

WENETSPEECH: A 10000+ Hours Multi-Domain Mandarin Corpus for Speech Recognition

GigaSpeech: An Evolving, Multi-domain ASR Corpus with 10, 000 Hours of Transcribed Audio

History Utterance Embedding Transformer LM for Speech Recognition

AISHELL-4: An Open Source Dataset for Speech Enhancement, Separation, Recognition and Speaker Diarization in Conference Scenario

Bayesian HMM clustering of x-vector sequences (VBx) in speaker diarization: theory, implementation and analysis on standard tasks

Conformer: Convolution-augmented Transformer for Speech Recognition

CHiME-6 Challenge: Tackling Multispeaker Speech Recognition for Unsegmented Recordings

Online Hybrid CTC/Attention End-to-End Automatic Speech Recognition Architecture

CN-Celeb: A Challenging Chinese Speaker Recognition Dataset

SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition

Dialog-Context Aware end-to-end Speech Recognition

TED-LIUM 3: twice as much data and corpus repartition for experiments on speaker adaptation

Speech-Transformer: A No-Recurrence Sequence-to-Sequence Model for Speech Recognition

ESPnet: End-to-End Speech Processing Toolkit

Additive Margin Softmax for Face Verification

Hybrid CTC/Attention Architecture for End-to-End Speech Recognition

AISHELL-1: An open-source Mandarin speech corpus and a speech recognition baseline

VoxCeleb: A Large-Scale Speaker Identification Dataset

Joint CTC-attention based end-to-end speech recognition using multi-task learning

Deep Residual Learning for Image Recognition

THCHS-30 : A Free Chinese Speech Corpus

Listen, attend and spell: A neural network for large vocabulary conversational speech recognition

Librispeech: An ASR corpus based on public domain audio books

Towards End-To-End Speech Recognition with Recurrent Neural Networks

On Speech Recognition

Recognition and understanding of meetings the AMI and AMIDA projects

The CHIL audiovisual corpus for lecture and meeting analysis inside smart rooms

HKUST/MTS: A Very Large Scale Mandarin Telephone Speech Corpus

Building multiclass classifiers for remote homology detection and fold recognition

Probabilistic Linear Discriminant Analysis

The Fisher Corpus: a Resource for the Next Generations of Speech-to-Text

The ICSI Meeting Corpus

SWITCHBOARD: telephone speech corpus for research and development

The Design for the Wall Street Journal-based CSR Corpus

Keyword Search Using Attention-Based End-to-End ASR and Frame-Synchronous Phoneme Alignments

The Kaldi Speech Recognition Toolkit

Connectionist Temporal Classiﬁcation: Labelling Unsegmented Sequence Data with Recurrent Neural Networks

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names