|
--- |
|
license: cc-by-nc-nd-4.0 |
|
--- |
|
## Description |
|
It collects 312 speakers from northeast regione. All speakers are reading the texts in northeast dialect. The recording contents cover customer consultant and message text from nearly 30 fields. Sentences are manually transcribed and proofread by professional annotators, with high accuracy. |
|
|
|
For more details, please refer to the link: https://www.nexdata.ai/dataset/49?source=Huggingface |
|
|
|
|
|
## Format |
|
16kHz, 16bit, uncompressed wav, mono channel |
|
|
|
## Recording Environment |
|
quiet indoor environment, without echo |
|
|
|
## Recording Content |
|
customer consulting (covered 30 fields); text message |
|
|
|
## Speaker |
|
312 people from Northeast China, 70% of which are female |
|
|
|
## Device |
|
Android mobile phone |
|
|
|
## Language |
|
Northeast dialect |
|
|
|
## Transcription content |
|
text |
|
|
|
## Accuracy rate |
|
95% |
|
|
|
## Application scenarios |
|
speech recognition, voiceprint recognition |
|
|
|
# Licensing Information |
|
Commercial License |
|
|
