Acoustic modeling by phoneme templates and modified one-pass DP decoding for continuous speech recognition.

V. Ramasubramanian, Kaustubh Kulkarni, Bernhard Kaemmerer

Browse the full ICASSP paper archive.