Prosodic Features in User Response to a Speech Dialogue System

スポンサーリンク

概要

• 論文の詳細を見る

• This paper discusses prosodic features characteristic of English in user responses to a speech dialogue system and examines how this information can be extracted from the speechwave. Feeling that the ability to recognize sub-lexical information is important for speech recognition systems, we looked at the way users of English and Japanese speech dialogue systems modified their utterances when asked for an exact repetition. By comparing data from the two systems, we tried to identify those English prosodic features which have the greatest influence on meaning. We outline here the experiment performed and present the results of acoustic analysis.

• 一般社団法人情報処理学会の論文
• 1993-09-27

著者

• Hayamizu S

  Electrotechnical Lab.

• Tanaka K

  Electrotechnical Lab.

• Itou K.

  Electrotechnical Laboratory

• Hayamizu S

  Electrotechnical Laboratory

• Mayfield L.J.

  Electrotechnical Laboratory

関連論文

• Speech Enhancement Using Array Signal Processing Based on the Coherent-Subspace Method
• On the Distance Measures Using Dynamic Features of Speech Parameters (音声認識)
• Phonetic Label(APSEG)Sequence Search System and Acoustic-Phonetic Variation Clustering Based on a Network Representation (音声認識)
• Some Features of Acoustic-Phonetic Segments of Voiceless Plosives and Their Relation to Speech Context (音声認識)
• Automatic Labeling of Known Speech Samples Using a Rule-Based Network Representation and Segmentation Technique (音声認識)
• Some Features of Voiceless Plosives in Phonetically Balanced Word Sets (音声認識)
• Semiautomatic Labeling for Constructing a Phoneme Data Base System (音声認識)
• Demiphonemic Representation of Speech and Semiautomatic Labeling System (音声認識特集)
• Pseudo-Phoneme Extraction by Unequal Rates of Sampling and Matrix Quantization (音声認識特集)
• Prosodic Features in User Response to a Speech Dialogue System

スポンサーリンク