Speech recognition is a computationally demanding task, particularly the stage which uses Viterbi decoding for converting pre-processed speech data into words or sub-word units. Any device that can reduce the load on, for example, a PC’s processor, is advantageous. Hence we present FPGA implementations of the decoder based alternately on discrete and continuous hidden Markov models (HMMs) representing monophones, and demonstrate that the discrete version can process speech nearly 5,000 times real time, using just 12% of the slices of a Xilinx Virtex XCV1000, but with a lower recognition rate than the continuous implementation, which is 75 times faster than real time, and occupies 45% of the same device.
|Name||Lecture Notes in Computer Science|
|Conference||12th International Conference on Field-Programmable Logic and Applications, Sep 02-04, 2002. MONTPELLIER, France|
|Period||1/01/02 → …|
The original publication is available at www.springerlink.com