This paper describes incorporating discriminative features from a multi layer perceptron (MLP) into a state-of-the-art. Arabic broadcast data transcription ...

PDF | This paper describes incorporating discriminative features from a multi layer perceptron (MLP) into a state-of-the-art Arabic broadcast data.

Apr 21, 2011 · The study covers the two main directions along which the MLP features have evolved: the use of different input representations to the MLP and ...

In this paper the use of MLP features is evaluated in a large vocabulary continuous speech recognition task, exploring different types of MLP features and their ...

22 Citations · Transcribing broadcast data using MLP features · The efficient incorporation of MLP features into automatic speech recognition systems.

This paper describes incorporating discriminative features from a multi layer perceptron (MLP) into a state-of-the-art Arabic broadcast data transcription ...

We describe extensions and improvements to IBM's system for automatic transcription of broadcast news. The speech recognizer uses a total of 160 hours of ...

Transcribing broadcast data using MLP features. P. Fousek, L. Lamel, J. Gauvain. 2008, Interspeech. Efficient generation and use of MLP features for Arabic ...

To test this, we used PLP features to train MLPs for every combination of 4 network sizes (hidden layers of. 500, 1000, 2000 and 4000 units) and 4 training-set ...

Results on 5 hours of evaluation data from the. GALE project reveal that the MLP features consistently produce improvements in the range of 15%–23% relative at ...