LABORATORIO DE COMUNICACIÓN ORAL
ROBERT WAYNE NEWCOMB
P. Gómez, A. Alvarez, R. Martínez, M. Pérez, V. Rodellar and V. Nieto
- "A Noise-Robust Speech Recogniser supported by a TMS320C31 Platform".
- The 2nd European DSP Education and Research Conference
- Paris, France, 23-24 de September, 1998
- Robust Speech Recognition Techniques are necessary in many applications where the environment is too noisy to obtain reliable recognition rates with standard Speech Recognisers. This is especially so in Car Cabinets, Avionics, Military Vehicles, Manufacturing Chains, Unreliable Communications Channels and others. The present paper is intended to show a Robust Isolated-Word Speech Recognition System working on a real-time platform based on the TMS320C31. The hardware platform developed for the project, and the set of noise cancelling and speech recognition applications produced are described. Results on the performance obtained with the noise-robust speech recogniser are also given and discussed.
Pulse aquí para bajarse el artículo
Formato PostScript (1.242 Kb.)
Comprimido en zip (580 Kb.)
Madrid a 9 de junio de 2004