Far-Field Speech Recognition with Real Reverberant Data
Speaker:Barbara Rauch
Institution: Saarland University
Abstract:
In this talk I will report on my recent work in the area of automatic speech recognition with microphone arrays. While my previous work mainly used artificially reverberated Wall Street Journal (WSJ) sentences, I will describe some recent experiments involving the use of the MC-WSJ-AV corpus, a database of real reverberant WSJ sentences recorded in the Edinburgh instrumented meeting room.