1. Core technologies for speech recognition:
New features for speech signals, new models and new frameworks for speech recognition, handling noise and channel effect, improved acoustic modeling and adaptation, improved language modeling and adaptation, spontaneous speech processing, Mandarin-English bilingual speech processing, prosody and tone modeling, etc.
2. Intelligent applications of speech recognition vender network environment:
Speech understanding, spoken dialog modeling and systems, semantic analysis, voice-based information retrieval, speech information summarization and distillation, spoken document understanding and organization, spoken key term extraction, speech synthesis, distributed speech processing technologies, etc.