source: gs3-installations/mars/trunk/sites/mars/collect/amc-essentia/essentia-2013-2014.profile.in@ 36966

Last change on this file since 36966 was 34770, checked in by davidb, 3 years ago

A template version used to by AUDIO-FRAME-TO-ESSENTIA-CSV-FEATURE-FILE.pl to generate a specific essentia-2013-2014.profile to process a particular 6-sec time-slice of the audio file

File size: 1.4 KB
Line 
1# The Deam 2013-2014 dataset are all 45 second samples
2# Looking to match with up with Grekow's results with Weka
3# using his private dataset where users rated Arousal
4# and Valence after listening to 6 seconds of music
5
6# ****
7# ****
8# ****
9# ****
10# The following needs to be dynamically set for the AMC dataset
11# ASSERT: (endTime - startTime) = 6 secs, just like the DEAM dataset
12startTime: @startTime@
13endTime: @endTime@
14
15# ****
16# ****
17# ****
18# No need for Frame output in AMC dataset
19##outputFrames: 1
20
21outputFormat: json
22
23#requireMbid: false
24
25# Standardize on 44.1 kHz as a sample rate
26analysisSampleRate: 44100.0
27
28lowlevel:
29# 2048 and 1024 are the extractors reported default values
30# frameSize: 2048
31# hopSize: 1024
32
33# Considered working with a slightly bigger framesize
34# frameSize: 4096
35# hopSize: 2048
36
37# Ultimately decided to work with slightly bigger framesize
38# align with 100msec intervales to better mesh with DEAM
39# ground-truth data
40#
41# => At 44.1 kHz analysisSampleRate, a hop size of 2205 produces
42# a frame of data for every 0.1 sec of audio
43 frameSize: 4096
44 hopSize: 2205
45
46
47# The following are some example values from the Essentia web page on extractor CLI tool
48# zeroPadding: 0
49# windowType: blackmanharris62
50# silentFrames: noise
51# stats: ["mean", "var", "median", "min", "max", "dmean", "dmean2", "dvar", "dvar2"]
Note: See TracBrowser for help on using the repository browser.