Neuro-inspired Speech Recognition with Recurrent Spiking Neurons

Ghani, Arfan; McGinnity, T.M.; Maguire, Liam; Harkin, Jim

doi:10.1007/978-3-540-87536-9_53

Cited by 22 publications

(16 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several studies have used reservoir-based approaches to perform spoken digit classification [21,22,23,24,25]. See [26] for a review of reservoir-based approaches in general.…”

Section: Reservoir-based Approachesmentioning

confidence: 99%

“…In most studies, the state of a spiking reservoir is low-pass filtered [21,22] before being sent to the readout layer. This allows the readout layer to be rapidly trained using any traditional non-spiking method for a single-layer architecture [25]. Zhang et al [27] is the only study that presented an LSM in which the readout layer was trained using a bio-inspired, spike-based learning rule.…”

Section: Reservoir-based Approachesmentioning

confidence: 99%

See 1 more Smart Citation

A spiking network that learns to extract spike signatures from speech signals

Tavanaei

Maida

2017

Neurocomputing

View full text Add to dashboard Cite

Spiking neural networks (SNNs) with adaptive synapses reflect core properties of biological neural networks. Speech recognition, as an application involving audio coding and dynamic learning, provides a good test problem to study SNN functionality. We present a simple, novel, and efficient nonrecurrent SNN that learns to convert a speech signal into a spike train signature. The signature is distinguishable from signatures for other speech signals representing different words, thereby enabling digit recognition and discrimination in devices that use only spiking neurons. The method uses a small, nonrecurrent SNN consisting of Izhikevich neurons equipped with spike timing dependent plasticity (STDP) and biologically realistic synapses. This approach introduces an efficient and fast network without error-feedback training, although it does require supervised training. The new simulation results produce discriminative spike train patterns for spoken digits in which highly correlated spike trains belong to the same category and low correlated patterns belong to different categories. The proposed SNN is evaluated using a spoken digit recognition task where a subset of the Aurora speech dataset is used. The experimental results show that the network performs well in terms of accuracy rate and complexity.

show abstract

“…Several studies have used reservoir-based approaches to perform spoken digit classification [21,22,23,24,25]. See [26] for a review of reservoir-based approaches in general.…”

Section: Reservoir-based Approachesmentioning

confidence: 99%

Section: Reservoir-based Approachesmentioning

confidence: 99%

A spiking network that learns to extract spike signatures from speech signals

Tavanaei

Maida

2017

Neurocomputing

View full text Add to dashboard Cite

show abstract

“…It was analysed in a previously published paper that small columns are more predictable and stable and hence can be used for parallel implementations [6].…”

Section: Reconfigurable Architecturementioning

confidence: 99%

Neuro inspired reconfigurable architecture for hardware/software co-design

Ghani

McDaid

Belatreche

et al. 2009

2009 IEEE International SOC Conference (SOCC)

Self Cite

View full text Add to dashboard Cite

In neuro-inspired computing, cortical columns or microcircuits appear to be the basic unit of computation for recognition related tasks. This paper proposes an integrated framework for an efficient implementation of a cortical column on reconfigurable platform (FPGAs). A novel hardware/software co-design implementation strategy is devised and the functionality is tested in the context of speech recognition application. An overall accuracy of 98% was achieved with the fixed-point hardware/software co-design technique in comparison to the 100% with the floating-point software implementation.

show abstract

“…Speaker independent isolated speech recognition experiments are conducted with the TI 46-Word database [27] that has been used in recent works like [28], [29]. The systems are developed on a desktop computer (Intel Core2 Duo with Windows Vista) under Matlab R2006a exploiting the Voicebox [25] and H2M [26] toolbox to support the MFCCs computations and HMMs respectively.…”

Section: A Experimental Setupmentioning

confidence: 99%

Variable frame rate hierarchical analysis for robust speech recognition

Rouat

Loiselle

Molotchnikoff³

2011

2011 IEEE/RSJ International Conference on Intelligent Robots and Systems

View full text Add to dashboard Cite

A new bio-inspired speech analysis system that extracts acoustical speech events is proposed and used in the design of a variable frame rate (VFR) speech recognizer. The same speech recognizer (Hidden Markov Model-HMMand Mel Frequency Cepstrum Coefficients-MFCC-) has been used with the proposed VFR analysis and conventional fixed frame rate (FFR) approach. In comparison with other VFR recognizers, the hierarchical features in the proposed system have the potential to serve as classification parameters of a complete bio-inspired speech recognition system. Also, no voice activity detection is required and there are no hard decisions to be taken by the system. Events are used to label and identify the moments at which the acoustical properties of speech are stable or changing. These events are markers on which an analysis window can be positioned to perform the recognition. Inspired by our knowledge of the auditory and visual systems, hierarchical complex features like transients and energy orientation are used. Training has been done on clean speech and recognition on noisy (from 20dB to-10dB Signal to Noise Ratios-SNR) or reverberated speech by using the TI 46-word database corrupted with 4 noises taken from the Aurora 2 data. In comparison with a FFR recognizer, our VFR system yields more than 50% increase in recognition rates for a speaker independent isolated word recognition task when SNRs are between 0 and 20 dB.

show abstract

Neuro-inspired Speech Recognition with Recurrent Spiking Neurons

Cited by 22 publications

References 13 publications

A spiking network that learns to extract spike signatures from speech signals

A spiking network that learns to extract spike signatures from speech signals

Neuro inspired reconfigurable architecture for hardware/software co-design

Variable frame rate hierarchical analysis for robust speech recognition

Contact Info

Product

Resources

About