Table 3 The contents of the two audio databases
From: An evolutionary feature synthesis approach for content-based audio retrieval
Class number | Audio class name | Number of samples | ||
|---|---|---|---|---|
12-class audio database (extended) | 6-class audio database (basic) | 1 | Female speech | 111 |
2 | Male singing | 63 | ||
3 | Whistling | 94 | ||
4 | Breaking glass | 83 | ||
5 | Classical music | 116 | ||
6 | Electrical/techno music | 107 | ||
The 6 added classes | 7 | Male speech | 103 | |
8 | Female singing | 119 | ||
9 | Bird singing | 91 | ||
10 | Dog barking | 49 | ||
11 | Fire sounds | 96 | ||
12 | Rock/metal music | 72 | ||
Total | 1104 | |||