Datasets:

Nexdata
/

mixed_speech_chinese_english

Tasks:

Automatic Speech Recognition

Modalities:

Formats:

Languages:

Size:

Libraries:

Dataset card Data Studio Files Files and versions Community

Nexdata commited on Apr 23

Commit

3a5ef02

·

verified ·

1 Parent(s): 95a0151

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -45,8 +45,8 @@ language:
 ### Dataset Summary
-The dataset contains 2,000 hours of mixed speech with Chinese and English. The data is collected from speakers in 26 provinces like Henan, Shanxi, Sichuan, Hunan, Fujian, etc.The content covers generic scene and multiple human machine interation scenes, such as music, entertainment, travel, daily life. The data covers more than 30,000 English words. The sentence accuracy is over 97%.
-For more details, please refer to the link: https://www.nexdata.ai/datasets/speechrecog?source=Huggingface
 ### Supported Tasks and Leaderboards

 ### Dataset Summary
+This dataset is just a sample of 2,000 hours of mixed speech with Chinese and English dataset(paid dataset). The data is collected from speakers in 26 provinces like Henan, Shanxi, Sichuan, Hunan, Fujian, etc.The content covers generic scene and multiple human machine interation scenes, such as music, entertainment, travel, daily life. The data covers more than 30,000 English words. The sentence accuracy is over 97%.
+ For more details & to download the rest of the dataset(paid),please refer to the link: https://www.nexdata.ai/datasets/speechrecog?source=Huggingface
 ### Supported Tasks and Leaderboards