Product Overview

Product Overview

Speech recognition is a service that accurately recognizes speech as text, enriching users’ communication methods. Combined with natural language processing (NLP), user intentions can be derived from speech, providing a basis for developers to formulate operational strategies.

Product Features

  • Supports multiple languages: able to recognize more than 20 languages including English, Japanese, Korean, Chinese, Indonesian, Filipino, Thai, Vietnamese, Arabic, Portuguese, Spanish, and Turkish.
  • Supports custom hot words: You can upload a personalized word list to transcribe terms and uncommon words in specific fields, and improve the recognition accuracy of specific words or phrases.
  • Anti-noise interference: Able to handle noisy audio from a variety of environments without the need for additional noise reduction measures.

Application Scenes

  • Supports accurate recognition of short audio (≤60 seconds) into text and returns results in real time
  • Supports converting long audio (<5 hours) into text data, providing a basis for text mining
  • Supports speaker differentiation (Chinese only), providing a basis for customer service quality inspection