WayWithWords

/

South African English Speech Collection Dataset

0

Data Summary

The dataset consists of 50 hours of simulated, unscripted agent-caller dialogue across domains like Insurance, Retail, Debt Collection, and Travel in South African English. It features 63 participants from various South African provinces with diverse educational backgrounds and qualifications.

Use Cases

  • Speech Recognition
  • Machine Learning (ML)
  • Natural Language Processing (NLP)