TingChen-ppmc's picture
Update README.md
f10f366 verified
metadata
dataset_info:
  features:
    - name: audio
      dtype: audio
    - name: gender
      dtype: string
    - name: speaker_id
      dtype: string
    - name: transcription
      dtype: string
  splits:
    - name: train
      num_bytes: 422057259.808
      num_examples: 3792
  download_size: 436738370
  dataset_size: 422057259.808
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Corpus

This dataset is built from Magicdata ASR-CZDIACSC: A CHINESE SHANGHAI DIALECT CONVERSATIONAL SPEECH CORPUS

This corpus is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Please refer to the license for further information.

Modifications: The audio is split in sentences based on the time span on the transcription file. Sentences that span less than 1 second is discarded. Topics of conversation is removed.

Usage

To load this dataset, use

from datasets import load_dataset
dialect_corpus = load_dataset("TingChen-ppmc/Shanghai_Dialect_Conversational_Speech_Corpus")

This dataset only has train split. To split out a test split, use

from datasets import load_dataset
train_split = load_dataset("TingChen-ppmc/Shanghai_Dialect_Conversational_Speech_Corpus", split="train")
# where test_size=0.5 denotes 0.5 of the dataset will be split to test split
corpus = train_split.train_test_split(test_size=0.5)

A sample data would be

# note this data is from the Nanchang Dialect corpus, the data format is shared
{'audio': 
    {'path': 'A0001_S001_0_G0001_0.WAV',
       'array': array([-0.00030518, -0.00039673, 
                     -0.00036621, ..., 	-0.00064087,
                      -0.00015259, -0.00042725]),
       'sampling_rate': 16000},
 'gender': '女',
 'speaker_id': 'G0001',
 'transcription': '北京爱数智慧语音采集'
}

More Information needed