Collecting supervised training data for automatic speech recognition (ASR) systems is both time consuming and expensive. In this paper we use the notion of virtual evidence in a g...
We present a data-mining experiment on feature selection for automatic emotion recognition. Starting from more than 1000 features derived from pitch, energy and MFCC time series, ...
Abstract. We apply Long Short-Term Memory (LSTM) recurrent neural networks to a large corpus of unprompted speech- the German part of the VERBMOBIL corpus. Training first on a fra...
Nicole Beringer, Alex Graves, Florian Schiel, J&uu...
In thispaper, we present a novelhybridarchitecture forcontinuousspeech recognition systems. It consists of a continuous HMM system extended by an arbitrary neural network that is ...
We have built a set of integrated AI systems (called conversation machines) to enable transaction processing over the telephone for limited domains like stock trading and banking....
Wlodek Zadrozny, Catherine G. Wolf, Nanda Kambhatl...