Japanese native speech dataset

Infomation

Name: Object teaching dataset
The Object teaching dataset consists of 70 spoken sentences combined 26 words of 26 Japanese phonemes, e.g., "これはおもちゃ" {kore wa omocha} in English "This is a toy." and "やわらかいよ" {yawarakai yo} in English "It's soft.". Each sentence is spoken once by a native Japanese speaker and recorded in the dataset.

File list

• DATA/ (Speech data)
• PHONELABEL/ (Phoneme labels)
• WORDLABEL/ (Word labels)
• utterance_list.txt (List of file names and sentences)

License

• MIT
  • see LICENSE