FLEXIBLE SPEECH UNDERSTANDING BASED ON COMBINED KEY-PHRASE DETECTION AND VERIFICATION

We propose a novel speech understanding strategy based on combined det ection and verification of semantically tagged key-phrases in spontane ous spoken utterances. Key-phrases are defined in a top-down manner so as to constitute semantic slots. Their detection directly leads to ro bust understanding. A phrase network realizes both a wide coverage and a reasonable constraint for detection. A subword-based verifier is th en incorporated to reduce false alarms in detection and attach confide nce measures of the detected phrases. This set of phrase confidence me asures, when incorporated in a spoken dialogue system, forms a basis f or designing intelligent speech interfaces that accept only verified k ey-phrases and reprompt users to clarify unspecified or unrecognized p ortions. Several forms of confidence measures based on subword-level t ests are investigated. The proposed approach was tested on field data collected from real-world trial applications. The combined detection a nd verification strategy drastically improves the accuracy in handling out-of-grammar utterances over the conventional decoding approaches w hile maintaining the performance for in-grammar utterances.