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

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/014yck   Aircraft engine
2    /t/dd00032   Funny music
3    /m/01280g   Wild animals
4    /m/027m70_   Jingle bell
5    /m/02qmj0d   String section
6    /m/01hsr_   Sneeze
7    /m/03w41f   Church bell
8    /m/01bns_   Zither
9    /m/07p6mqd   Slosh
10    /m/0319l   French horn
11    /m/02_nn   Fart
12    /m/07qfgpx   Jingle- tinkle
13    /m/07r10fb   Raindrop
14    /m/02bxd   Didgeridoo
15    /m/06_y0by   Environmental noise
16    /m/03qc9zr   Screaming
17    /m/078jl   Snake
18    /m/09ld4   Frog
19    /m/07qz6j3   Whimper
20    /m/0jb2l   Thunderstorm
21    /m/02pjr4   Blender
22    /m/01jnbd   Echo
23    /m/0ngt1   Thunder
24    /m/07cx4   Telephone
25    /g/122z_qxw   Firecracker
26    /m/03q5_w   Burping- eructation
27    /m/07p_0gm   Throbbing
28    /m/03xq_f   Electronic organ
29    /m/07p9k1k   Sizzle
30    /m/07pqc89   Trickle- dribble
31    /m/07qmpdm   Clatter
32    /m/011k_j   Timpani
33    /m/07p6fty   Shout
34    /m/03gvt   Hammond organ
35    /m/07rqsjt   Whoosh- swoosh- swish
36    /m/03r5q_   Jingle (music)
37    /m/07rcgpl   Hum
38    /m/0b_fwt   Electronic tuner
39    /m/02_41   Fire
40    /t/dd00031   Happy music
41    /m/0463cq4   Crying- sobbing
42    /m/07pjjrj   Smash- crash
43    /m/07qfr4h   Hubbub- speech noise- speech babble
44    /m/07qwyj0   Rustle
45    /m/0j2kx   Waterfall
46    /m/07pzfmf   Crackle
47    /m/0dwt5   Vibraphone
48    /m/07rknqz   Skidding
49    /m/0h2mp   Fly- housefly
50    /m/0130jx   Sink (filling or washing)