A5033838135- Speech Recognition and Synthesis
- Speech and Audio Processing
- Advanced Data Compression Techniques
- Speech and dialogue systems
- Biomedical Text Mining and Ontologies
- Topic Modeling
- Intellectual Property and Patents
- Semantic Web and Ontologies
- Wireless Communication Networks Research
- Cloud Data Security Solutions
- Blockchain Technology Applications and Security
- Music and Audio Processing
- Private Equity and Venture Capital
- Blind Source Separation Techniques
- Scientific Computing and Data Management
- Advanced Adaptive Filtering Techniques
- Advanced Computational Techniques and Applications
- Service-Oriented Architecture and Web Services
- Algorithms and Data Compression
- Big Data and Business Intelligence
Ospedale di Rivoli
2021
Intel (United States)
2019
Telecom Italia Lab
1985-2005
The development of spoken dialogue systems (SDSs) requires the definition evaluation metrics which can assess performance these at different levels and compare various SDSs. authors present a first test, made with naive users, on an integrated system for telephone speech access to remote data base. They describe architecture as well goals its features, methodology used during evaluation, results obtained. SDS is shown be effective providing user required information. presence spontaneous...
The paper is a survey of the speech technologies and applications developed at CSELT, some which are employed in real services on Italian telephone network. This represents major extension from previous activity centre, was essentially oriented to research, now broader set activities range defining experimenting new algorithmic approaches product engineering application development. In particular describes two field trial phase, one related an automatic operator, providing voice selection...
Here we describe the firmware implementation of an acoustical front-end, performing vector quantization Discrete Cosine Transform (DCT) for a speech recognition system. This runs on single TMS32020 signal processor chip and is characterized both by substantial real time performance good accuracy.
First Page