| Title: | Hardware–software co-design of an audio feature extraction pipeline for machine learning applications |
|---|
| Authors: | ID Vreča, Jure, Institut Jožef Stefan (Author)
ID Pilipović, Ratko (Author)
ID Biasizzo, Anton, Institut Jožef Stefan (Author) |
|---|
| Files: |  URL - Source URL, visit https://www.mdpi.com/2079-9292/13/5/875
 PDF - Presentation file, download (1,05 MB)
MD5: F09ED506C9C50118B6B65AC40E9F9684
|
|---|
| Language: | English |
|---|
| Typology: | 1.01 - Original Scientific Article |
|---|
| Organization: |  IJS - Jožef Stefan Institute
|
|---|
| Abstract: | Keyword spotting is an important part of modern speech recognition pipelines. Typical contemporary keyword-spotting systems are based on Mel-Frequency Cepstral Coefficient (MFCC) audio features, which are relatively complex to compute. Considering the always-on nature of many keyword-spotting systems, it is prudent to optimize this part of the detection pipeline. We explore the simplifications of the MFCC audio features and derive a simplified version that can be more easily used in embedded applications. Additionally, we implement a hardware generator that generates an appropriate hardware pipeline for the simplified audio feature extraction. Using Chisel4ml framework, we integrate hardware generators into Python-based Keras framework, which facilitates the training process of the machine learning models using our simplified audio features. |
|---|
| Keywords: | FPGA, MFCC, keyword spotting, chisel |
|---|
| Publication status: | Published |
|---|
| Publication version: | Version of Record |
|---|
| Submitted for review: | 31.01.2024 |
|---|
| Article acceptance date: | 22.02.2024 |
|---|
| Publication date: | 24.02.2024 |
|---|
| Publisher: | MDPI |
|---|
| Year of publishing: | 2024 |
|---|
| Number of pages: | str. 1-14 |
|---|
| Numbering: | 13, 5 |
|---|
| Source: | Švica |
|---|
| PID: | 20.500.12556/DiRROS-18556  |
|---|
| UDC: | 004 |
|---|
| ISSN on article: | 2079-9292 |
|---|
| DOI: | 10.3390/electronics13050875 |
|---|
| COBISS.SI-ID: | 186803203 |
|---|
| Copyright: | © 2024 by the authors. |
|---|
| Note: | Nasl. z nasl. zaslona;
Opis vira z dne 26. 2. 2024;
|
|---|
| Publication date in DiRROS: | 25.03.2024 |
|---|
| Views: | 582 |
|---|
| Downloads: | 502 |
|---|
| Metadata: |    |
|---|
|
:
|
Copy citation |
|---|
| | | | Share: |  |
|---|
Hover the mouse pointer over a document title to show the abstract or click
on the title to get all document metadata. |
