KNOWLEDGE TRANSFER FROM WEAKLY LABELED AUDIO USING CONVOLUTIONAL NEURAL NETWORK FOR SOUND EVENTS AND SCENES. pdf
Authors: Anurag Kumar, Maksim Khadkevich, Christian Fügen
Area Under ROC Curves (AUC)
![average Precision](./auc2.png)
Figure compared AP of \(\mathcal{N}_S\) and \(\mathcal{N}_S^{slat}\). The Index of sound events 1 to 50 is shown below. The first column is Index in the above figure, second is the event id (magenta) as used in Audioset dataset and third (blue) is sound event name
1 /m/096m7z Noise
2 /m/0dl9sf8 Throat clearing
3 /m/07m2kt Chorus effect
4 /m/07pws3f Bang
5 /m/07hvw1 Field recording
6 /m/024dl Cash register
7 /m/07pdhp0 Biting
8 /m/07rn7sz Shatter
9 /m/01s0vc Zipper
10 /m/07qcx4z Tearing
11 /m/015jpf Dial tone
12 /m/07qlwh6 Squish
13 /m/07s12q4 Crunch
14 /m/07st88b Croak
15 /m/02rhddq Reversing beeps
16 /m/03wvsk Hair dryer
17 /m/0242l Coin
18 /m/07prgkl Pour
19 /m/05r5wn Rattle
20 /m/07qb_dv Scratch
21 /m/025wky1 Air conditioning
22 /m/04rmv Mouse
23 /t/dd00088 Gush
24 /m/07qv4k0 Scrape
25 /m/02fxyj Humming
26 /m/0dv5r Camera
27 /t/dd00065 Light engine
28 /m/07pczhz Chop
29 /m/07rrh0c Thunk
30 /m/0k65p Hands
31 /m/06xkwv Mains hum
32 /m/01jg1z Heart murmur
33 /m/04gy_2 Battle cry
34 /m/0dv3j Boiling
35 /m/07rbp7_ Whip
36 /m/07pjwq1 Buzz
37 /m/01v_m0 Sine wave
38 /m/02c8p Telephone dialing
39 /m/07pc8lb Breaking
40 /m/07pyf11 Flap
41 /m/07r67yg Ding-dong
42 /t/dd00121 Boing
43 /m/07bjf Single-lens reflex camera
44 /m/0mbct Gong
45 /m/01swy6 Yodeling
46 /m/07r81j2 Caterwaul
47 /m/07phhsh Rumble
48 /t/dd00048 Train wheels squealing
49 /m/03l9g Hammer
50 /m/03v3yw Keys jangling