Next Article in Journal
Edge Computing Robot Interface for Automatic Elderly Mental Health Care Based on Voice
Next Article in Special Issue
A Fully-Integrated Analog Machine Learning Classifier for Breast Cancer Classification
Previous Article in Journal
Analytical and Experimental Evaluation of a Novel Mechanism to Improve the Control Plane in Next-Generation Mobile Networks
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Electronics
Volume 9
Issue 3
10.3390/electronics9030418
Review Report
electronics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
Article
Peer-Review Record

Time-Encoding-Based Ultra-Low Power Features Extraction Circuit for Speech Recognition Tasks

Electronics 2020, 9(3), 418; https://doi.org/10.3390/electronics9030418
by Eric Gutierrez *, Carlos Perez, Fernando Hernandez and Luis Hernandez
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Electronics 2020, 9(3), 418; https://doi.org/10.3390/electronics9030418
Submission received: 30 January 2020 / Revised: 26 February 2020 / Accepted: 26 February 2020 / Published: 29 February 2020
(This article belongs to the Special Issue Analog and Digital Circuit Design Techniques and Systems for Machine Learning)

Round 1

Reviewer 1 Report

Dear authors,

 

thank you for submitting a paper of good quality. I feel that in the table of comparison, a few typical VCO figures of merit would even improve the paper. I am particularly thinking of the VCO jitter and PSRR. Especially since you are using single ended inverters. Perhaps you could motivate this choice rather than fully-diff delay cells.

 

Best regards

Author Response

Please, you can find the answers in the attached .pdf file. Thanks.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper proposes a new approach for circuits devoted to speech recognition using a bank a band-pass filters implemented with ring-oscillators to extract the features (energy within different frequency bands) of input audio signals and digitize them.

Comments:

  1. Improve the title. Currently, it is confusing. Avoid abbreviations.
  2. Why you selected 65-nm CMOS process technology to demonstrate application? This technology is 15 year old.
  3. The process of estimating power consumption should be described in more detail. Discuss various methods to perform such estimation, for example, at the behavioral level, see: DOI: 10.1155/2007/68673, or statistical method, see doi: 10.1007/s11277-018-5927-7
  4. Table 1 does not provide for power consumption.
  5. Check the numbering of Tables (Table II vs Table 2).
  6. Discuss the limitations of your approach and threats-to-validity of experimental results.
  7. Improve conclusions; support your claims with a summary of experimental results.

Author Response

Please, find the answers in the attached .pdf file. Thanks.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors have addressed all my remarks. I suggest to check carefully all figures and explain all abbreviations used (such as PMOS, NMOS) in respective figure captions.

Author Response

Thanks for your feedback.

We have checked all the figures and we have added the missing abbreviations to the list of abbreviations at the end of the document.

Back to TopTop