Download now Free registration required
Effective human and automatic processing of speech requires recovery of more than just the words. It also involves recovering phenomena such as sentence boundaries, filler words, and disfluencies, referred to as structural metadata. The paper describes a metadata detection system that combines information from different types of textual knowledge sources with information from a prosodic classifier. The paper investigates maximum entropy and conditional random field models, as well as the predominant HMM approach, and find that discriminative models generally outperform generative models. The paper reports system performance on both broadcast news and conversational telephone speech tasks, illustrating significant performance differences across tasks and as a function of recognizer performance. The results represent the state of the art, as assessed in the NIST RT-04F evaluation.
- Format: PDF
- Size: 256 KB