eswardivi/IndicVoices

--- dataset_info: - config_name: telugu_train features: - name: audio dtype: audio - name: end dtype: float64 - name: text dtype: string - name: start dtype: float64 - name: speaker_id dtype: int64 - name: audio_raw_duration dtype: float64 - name: scenario dtype: string - name: task_name dtype: string splits: - name: split1 num_bytes: 1609730034.0 num_examples: 1000 - name: split2 num_bytes: 1049782097.0 num_examples: 1000 - name: split3 num_bytes: 1444386250.0 num_examples: 1000 - name: split4 num_bytes: 3152422807.0 num_examples: 1000 - name: split5 num_bytes: 5554666300.0 num_examples: 1000 - name: split6 num_bytes: 4003049531.0 num_examples: 500 - name: split7 num_bytes: 4053228693.0 num_examples: 500 - name: split8 num_bytes: 1044856061.0 num_examples: 344 - name: split9 num_bytes: 5059082700.0 num_examples: 500 - name: split10 num_bytes: 4664957526.0 num_examples: 500 download_size: 29946910673 dataset_size: 31636161999.0 - config_name: telugu_valid features: - name: audio dtype: audio - name: text dtype: string - name: audio_raw_duration dtype: float64 - name: scenario dtype: string - name: task_name dtype: string splits: - name: valid num_bytes: 314053800.0 num_examples: 73 download_size: 301914962 dataset_size: 314053800.0 configs: - config_name: telugu_train data_files: - split: split1 path: telugu_train/split1-* - split: split2 path: telugu_train/split2-* - split: split3 path: telugu_train/split3-* - split: split4 path: telugu_train/split4-* - split: split5 path: telugu_train/split5-* - split: split6 path: telugu_train/split6-* - split: split7 path: telugu_train/split7-* - split: split8 path: telugu_train/split8-* - split: split9 path: telugu_train/split9-* - split: split10 path: telugu_train/split10-* - config_name: telugu_valid data_files: - split: valid path: telugu_valid/valid-* --- This dataset is processed from [indicvoices](https://ai4bharat.iitm.ac.in/indicvoices/)

原始信息汇总

数据集概述

配置名称：telugu_train

• 特征：
  • audio: 音频
  • end: 浮点数
  • text: 字符串
  • start: 浮点数
  • speaker_id: 整数
  • audio_raw_duration: 浮点数
  • scenario: 字符串
  • task_name: 字符串
• 分割：
  • split1: 1609730034.0 字节, 1000 个样本
  • split2: 1049782097.0 字节, 1000 个样本
  • split3: 1444386250.0 字节, 1000 个样本
  • split4: 3152422807.0 字节, 1000 个样本
  • split5: 5554666300.0 字节, 1000 个样本
  • split6: 4003049531.0 字节, 500 个样本
  • split7: 4053228693.0 字节, 500 个样本
  • split8: 1044856061.0 字节, 344 个样本
  • split9: 5059082700.0 字节, 500 个样本
  • split10: 4664957526.0 字节, 500 个样本
• 下载大小：29946910673 字节
• 数据集大小：31636161999.0 字节

配置名称：telugu_valid

• 特征：
  • audio: 音频
  • text: 字符串
  • audio_raw_duration: 浮点数
  • scenario: 字符串
  • task_name: 字符串
• 分割：
  • valid: 314053800.0 字节, 73 个样本
• 下载大小：301914962 字节
• 数据集大小：314053800.0 字节

数据文件路径

• telugu_train：
  • split1: telugu_train/split1-*
  • split2: telugu_train/split2-*
  • split3: telugu_train/split3-*
  • split4: telugu_train/split4-*
  • split5: telugu_train/split5-*
  • split6: telugu_train/split6-*
  • split7: telugu_train/split7-*
  • split8: telugu_train/split8-*
  • split9: telugu_train/split9-*
  • split10: telugu_train/split10-*
• telugu_valid：
  • valid: telugu_valid/valid-*