Datasets:

Modalities:
Audio
Text
Languages:
English
Size:
< 1K
Libraries:
Datasets
License:
karenau commited on
Commit
34aafb1
·
verified ·
1 Parent(s): cbbc6a8

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +21 -3
README.md CHANGED
@@ -1,3 +1,21 @@
1
- ---
2
- license: cc-by-4.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-4.0
3
+ language:
4
+ - en
5
+ tags:
6
+ - audio
7
+ - speech
8
+ ---
9
+
10
+
11
+ ### Dataset Description
12
+
13
+ This dataset consists of 50 speech audio files, each sentence separated into their own .wav files, and a text file that consists of the transcript of all 50 sentences. The sentences were all taken from the book The Ballad of Songbirds and Snakes by Suzanne Collins. The audio files were recorded via Praat using the speakers from a Macbook Air in a quiet room.
14
+
15
+
16
+
17
+ ### Issues Encountered and Solutions
18
+
19
+
20
+
21
+ One of two issues I encountered was due to the lack of any microphones I had, I decided to record all the audio files on my phone. When I finished, I intended to transfer it to my laptop to work on, but despite numerous methods of transferral, all the files ended up corrupted with a glitch in the noise. I had to proceed to record it all again on my laptop, which unfortunately has subpar quality. Thankfully, the noise reduction helped reach the noise quality I had done on my phone. Another problem I encountered was getting really tired when reciting the lines. Because of this, the lines would often vary in loudness and pronunciation as I got more and more tired from recording, especially the final sentences towards the end. I opted to take breaks in between recording sessions and drink water beforehand to ensure that the quality of my voice could be ensured.