Continuous speech recognition for the TIMIT database using neural networks

Four types of neural networks which have previously been established for speech recognition and tested on a small, seven-speaker, 100-sentence database are applied to the TIMIT database. The networks are a recurrent network phoneme recognizer, a modified Kanerva model morph recognizer, a compositional representation phoneme-to-word recognizer, and a modified Kanerva model morph-to-word recognizer. The major result is for the recurrent net, giving a phoneme recognition accuracy of 57% from the si and sx sentences. The Kanerva morph recognizer achieves 66.2% accuracy for a small subset of the sa and sx sentences. The results for the word recognizers are incomplete.

Dynamic learning with the EM algorithm for neural networks

In this paper, we derive an EM algorithm for nonlinear state space models. We use it to estimate jointly the neural network weights, the model uncertainty and the noise in the data. In the E-step we apply a forwardbackward Rauch-Tung-Striebel smoother to compute the network weights. For the M-step, we derive expressions to compute the model uncertainty and the measurement noise. We find that the method is intrinsically very powerful, simple and stable.

Paraphrastic language models and combination with neural network language models

In natural languages multiple word sequences can represent the same underlying meaning. Only modelling the observed surface word sequence can result in poor context coverage, for example, when using n-gram language models (LM). To handle this issue, paraphrastic LMs were proposed in previous research and successfully applied to a US English conversational telephone speech transcription task. In order to exploit the complementary characteristics of paraphrastic LMs and neural network LMs (NNLM), the combination between the two is investigated in this paper. To investigate paraphrastic LMs' generalization ability to other languages, experiments are conducted on a Mandarin Chinese broadcast speech transcription task. Using a paraphrastic multi-level LM modelling both word and phrase sequences, significant error rate reductions of 0.9% absolute (9% relative) and 0.5% absolute (5% relative) were obtained over the baseline n-gram and NNLM systems respectively, after a combination with word and phrase level NNLMs. © 2013 IEEE.

In vitro recording of neural activity using carbon nanosheet microelectrodes

The advent of nanotechnology has revolutionised our ability to engineer electrode interfaces. These are particularly attractive to measure biopotentials, and to study the nervous system. In this work, we demonstrate enhanced in vitro recording of neuronal activity using electrodes decorated with carbon nanosheets (CNSs). This material comprises of vertically aligned, free standing conductive sheets of only a few graphene layers with a high surfacearea- to-volume ratio, which makes them an interesting material for biomedical electrodes. Further, compared to carbon nanotubes, CNSs can be synthesised without the need for metallic catalysts like Ni, Co or Fe, thereby reducing potential cytotoxicity risks. Electrochemical measurements show a five times higher charge storage capacity, and an almost ten times higher double layer capacitance as compared to TiN. In vitro experiments were performed by culturing primary hippocampal neurons from mice on micropatterned electrodes. Neurophysiological recordings exhibited high signal-to-noise ratios of 6.4, which is a twofold improvement over standard TiN electrodes under the same conditions. © 2013 Elsevier Ltd. All rights reserved.