In this paper we address the problem of building a good speech recognizer if there is only a small amount of training data available. The acoustic models can be improved by interpo...
Stefan Steidl, Georg Stemmer, Christian Hacker, El...
For extractive meeting summarization, previous studies have shown performance degradation when using speech recognition transcripts because of the relatively high speech recogniti...
Long-span features, such as syntax, can improve language models for tasks such as speech recognition and machine translation. However, these language models can be difficult to u...
We propose Cross-Channel Spectral Subtraction (CCSS), a source separation method for recognizing meeting speech where one microphone is prepared for each speaker. The method quick...
This paper first introduces a newly-recorded high quality Romanian speech corpus designed for speech synthesis, called “RSS”, along with Romanian front-end text processing mo...
Adriana Stan, Junichi Yamagishi, Simon King, Matth...