Scribd
Upload
0 views2 pages

ECE3001Proj Part1

The ECE3001 project focuses on speaker identification through voice data collection. Participants are required to record their voice for 10 minutes, convert it to WAV format, and segment the audio into 10-second files using provided code. Finally, the segmented files must be compressed into a ZIP file for submission on Blackboard.

Uploaded by

leoyuer631
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0 views2 pages

ECE3001Proj Part1

The ECE3001 project focuses on speaker identification through voice data collection. Participants are required to record their voice for 10 minutes, convert it to WAV format, and segment the audio into 10-second files using provided code. Finally, the segmented files must be compressed into a ZIP file for submission on Blackboard.

Uploaded by

leoyuer631
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

ECE3001 Project: Speaker Identification

Part I: Data Collection

1 Overview
The ECE3001 project is designed for speaker identification, which is a crucial task in the field of signal processing
and machine learning. Speaker identification aims to determine the identity of a person based on his/her speech
characteristics.

2 Procedures
1. Record your own voice on your phone for 10 minutes. You can read the material at the link TED talk at
a constant speed, in any language (Mandarin, English, dialect, etc.)
2. Convert the audio file to WAV format. It is recommended to use online video conversion tools like “迅捷
视频转换器”

3. Segment the audio file (the converted WAV format) into 10-second files by running the code provided
Split audio.py. Here is the detailed procedure:

(a) Make sure you have installed the required packages soundfile and librosa. If not, install them by
running the following command:
pip install soundfile librosa
(b) Put the audio file in the same directory as Split audio.py. Rename it to id<student id>.wav
(e.g., id219040062.wav). Open a terminal and navigate to this directory.
(c) Make a new folder at this directory to store the segmented audio files. The folder name is id<student id>
(e.g., id219040062).
(So far, you should have Split audio.py id<student id>.wav, and id<student id> folder in the
same directory.)
(d) Run the following command. Remember to replace ”<path to audio file>” with the actual path to
the audio file and replace ”<path to output directory>” with the directory you created in step 2.
python Split audio.py --in audio "<path to audio file>" --out dir "<path to output directory>"

1
For example, if your student id is 219040062, then the audio file should be id219040062.wav and
output folder should be id219040062. The command executed should be:
python Split audio.py --in audio "id219040062.wav" --out dir "id219040062"
(e) It will takes about 5-15 seconds to complete. Assume the student id is 219040062. Then the code
will read the audio file and save the 60 files in id219040062 folder as
• id219040062 01.wav
• ···
• id219040062 60.wav

4. Finally, compress the folder into a ZIP file named idstudent id.zip (i.e.,id219040062.zip) and submit
the zip file to the assignment link on the Blackboard.

You might also like