Pervasive Speech and Language Technology

Pervasive Speech and Language Technology

Pervasive Speech and Language Technology

Three Levels of Language Processing

Challenges for Language Engineering

Context-Sensitive Speech-to-Speech Translation

Mobile Speech-to-Speech Translation of Spontaneous Dialogs

Mobile Speech-to-Speech Translation of Spontaneous Dialogs

Speech-to-Speech Translation

The Control Panel of Verbmobil

General Speech Recognition Task

Word Hypotheses Graphs (WHGs)

Massive Data Collection Efforts

Extracting Statistical Properties from Large Corpora

From Multi-Agent Architectures to a Multi-Blackboard Architectures

A Multi-Blackboard Architecture for the Combination of Results from Deep and Shallow Processing Modules

The Use of Prosodic Information at All Processing Stages

Competing Strategies for Robust Speech Translation

Integrating Shallow and Deep Analysis Components in a Multi-Blackboard Architecture

VHG: A Packed Chart Representation of Partial Semantic Representations

The Understanding of Spontaneous Speech Repairs

Automatic Understanding and Correction of Speech Repairs in Spontaneous Telephone Dialogs

Robust Dialog Semantics: Combining and Completing Partial Representations

Integrating Deep and Shallow Processing: Combining Results from Concurrent Translation Threads

Unit Selection Algorithm

Linguatronic : Spoken Dialogs with Mercedes-Benz

International Research Trends in Multilingual Systems

Conclusion I

Conclusion II

Conclusion III

Open Problems for the Next Decade

A Speculative Conclusion (+50 years)