Energy-Efficient Floating-Point MFCC Extraction Architecture for Speech Recognition Systems

Jo, Jihyuck; Yoo, Hyung-Joun; Park, In-Cheol

doi:10.1109/tvlsi.2015.2413454

Cited by 48 publications

(31 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Inspired by the human hearing mechanisms, Mel-Frequency Cepstrum Coefficients (MFCC) feature is presented and becomes the most widely used feature [4] due to its high accuracy in this field. However, in the ASR tasks for mobile devices, the entire MFCC feature extraction process accounts for nearly 32% to 93% of system power consumption [3]- [5].…”

Section: (A)mentioning

confidence: 99%

“…Therefore, a lot of works have been continuously proposed to increase the efficiency of extracting MFCC feature. Fully considering the arithmetic property, Jo et al [4] proposed an energy-efficient floating-point MFCC extraction architecture based on field-programmable gate array (FPGA) with the improvement of frequency transformation and optimization of bit-width. Some other works [11], [12] about efficient MFCC extraction are also proposed based on FPGA for low-cost speech recognition systems.…”

Section: (A)mentioning

confidence: 99%

See 1 more Smart Citation

MSP-MFCC: Energy-Efficient MFCC Feature Extraction Method With Mixed-Signal Processing Architecture for Wearable Speech Recognition Applications

Yang

Lan

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Feature extraction is an essential part of automatic speech recognition (ASR) to compress raw speech data and enhance features, where conventional implementation methods based on the digital domain have encountered energy consumption and processing speed bottlenecks. Thus, we propose a Mixed-Signal Processing (MSP) architecture to efficiently extract Mel-Frequency Cepstrum Coefficients (MFCC) features. We design MSP-MFCC to pre-process speech signals in the analog domain, which significantly reduces the cost of the analog-to-digital converter (ADC), as well as the computational complexity of the digital backend. Moreover, MSP-MFCC eliminates the time-consuming Fourier transform in the conventional digital realization by improving processing flow. We fabricated the analog part based on 180nm CMOS mixedsignal technology, then measured the chip. The measured results show the energy consumption of MSP-MFCC is 0.72 µJ/frame, and the processing speed is up to 45.79 µs/frame. MSP-MFCC achieves 95% energy saving and about 6.4× speedup than state of the art. Further, by using the features extracted by MSP-MFCC, speech recognition simulation reaches the accuracy of 98.2%, which also keeps the leading performance to its current counterparts. The proposed MFCC extractor is competitive for integration in the ultra-low-power always-on wearable speech recognition applications. INDEX TERMS Mixed signal processing architecture, energy-efficient feature extraction, mel-frequency cepstrum coefficients (MFCC), wearable speech recognition application.

show abstract

Section: (A)mentioning

confidence: 99%

Section: (A)mentioning

confidence: 99%

MSP-MFCC: Energy-Efficient MFCC Feature Extraction Method With Mixed-Signal Processing Architecture for Wearable Speech Recognition Applications

Yang

Lan

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…As the MFCC closely represent a natural human hearing process, its coefficients are useful in speech classification of either speech recognition [17] or emotion recognition [18].…”

Section: Mel-frequency Cepstral Coefficientsmentioning

confidence: 99%

A Storyteller’s Tale: Literature Audiobooks Genre Classification Using CNN and RNN Architectures

Carmi¹,

Cohen²,

Avigal

et al. 2019

Interspeech 2019

View full text Add to dashboard Cite

Identifying acoustic properties that characterize reading literary genres can assist in giving a more personal and human tone to the speech of bots and automatic readings. In this paper we consider the following question: given speech segments of audiobooks, how well can we classify them according to their literary genres? In this study we consider three different literary genres: children, horror and suspense, and humorous audio books, taken from two free audio books sites: Librivox and YouTube. We ran four classification experiments: three for each pair of genres, and one for all three genres together. We repeated each experiment twice, with two different network architectures: Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). Note that, throughout the reading, there are sections that are more typical to the book's genre than others. As the samples were taken sequentially throughout the reading of the books and were short in duration, we did not expect high classification rates. Nevertheless, the accuracy of all the experiments were at least 72% for all the pair's classifications; and at least 57% for both architectures for the three classes classifications.

show abstract

“…The system can be scaled to include other health markers and can also be made userspecific.but the drawback was that it Consumes more number of hardware resources. [2] proposed architecture which employs floating-point arithmetic operations to minimize the operation bit-width and the total size of LUTs. Furthermore, a floating-point MAC unit and memories are shared with many processes to reduce hardware complexity and energy consumption remarkably but at the cost of operating Speed.…”

Section: Introductionmentioning

confidence: 99%

Design of a Low Power and Area Efficient Architecture for the Detection of Audio Biological Signals

Sangavi.¹

2018

IJRASET

View full text Add to dashboard Cite

In digital audio recording the audio signals are picked by a microphone or other transducer and converted into a stream of discrete numbers, representing the changes over time in air pressure for audio, then recorded to a storage device. Hence these audio recordings are analyzed to detect the audio biological symptoms, such as cough, sneeze, vomiting, wheezing, belching and so on, which are spectrally analyzed using a discrete wavelet transform (DWT). The DWT will help to find out the signal level of variation and also use simple mathematical metrics, such as energy, quasi-average, and coastline parameters. These parameters are used to find out type of symptomatic patterns to be detected. Furthermore a Mel-frequency cepstrum-based analysis is applied to distinguish between signals, such as cough and sneeze, which have similar frequency response and hence occur in common wavelet coefficients. The proposed approach is to detect the symptomatic patterns using acoustic non speech human signals which increases the efficiency of mathematical metrics and in particular reduces the area occupied by the architecture. Thus the aim of the proposed work is to design a low power and area efficient mathematical architecture for the calculation of energy parameter, coastline parameter, quasi-average and Mel Cepstrum based analysis for the detection of different symptomatic patterns in audio biological signals. Existing method uses Binary Common Sub-Expression Elimination (BCSE) technique in the design of multiplier which is used in the architecture of DWT and energy parameter calculation. The proposed work employs Multiple Constant Multiplier (MCM) technique in the design of DWT and energy parameter calculation. The design based on MCM outperforms the design using common sub-expression elimination technique with respect to hardware resources and power consumption thereby making the design a low power and area efficient.

show abstract

Energy-Efficient Floating-Point MFCC Extraction Architecture for Speech Recognition Systems

Cited by 48 publications

References 9 publications

MSP-MFCC: Energy-Efficient MFCC Feature Extraction Method With Mixed-Signal Processing Architecture for Wearable Speech Recognition Applications

MSP-MFCC: Energy-Efficient MFCC Feature Extraction Method With Mixed-Signal Processing Architecture for Wearable Speech Recognition Applications

A Storyteller’s Tale: Literature Audiobooks Genre Classification Using CNN and RNN Architectures

Design of a Low Power and Area Efficient Architecture for the Detection of Audio Biological Signals

Contact Info

Product

Resources

About