Title: Labelling Emotional User States in Speech: Where's the Problems, where's the Solutions?
1Labelling Emotional User States in
SpeechWhere's the Problems, where's the
Solutions?
- Anton Batliner, Stefan SteidlUniversity of
Erlangen - HUMAINE WP5-WS, Belfast, December 2004
2Overview
- decisions to be made
- mapping data onto labels
- the human factor
- and later on again some mapping
- illustrations
- the sparse data problem
- new dimensions
- new measure for labeller agreement
- and afterwards what to do with the data?
- handling of databases
- we proudly present CEICES
- some statements
3Mapping data onto labels I
- catalogue of labels
- data-driven (selection from "HUMAINE"-catalogue?)
- should be a semi-open class
- unit of annotation
- word (phrase or turn) in speech
- in video?
- alignment of video time stamps with speech data
necessary
4Mapping data onto labels II
- categorical (hard/soft) labelling vs. dimensions
- formal vs. functional labelling
- functional holistic user states
- formal prosody, voice quality, syntax, lexicon,
FAPs, ... - reference baseline
- speaker-/user-specific
- neutral phase at beginning of interaction
- sliding window
- emotion content vs. signs of emotion?
5The human factor I
- expert labellers vs. naïve labellers
- experts
- experienced, i.e. consistent
- with (theoretical) bias
- expensive
- few
- "naïve" labellers
- maybe less consistent
- no bias, i.e., ground truth?
- less expensive
- more
- representative data many data high effort
- are there "bad" labellers?
- does high interlabeller agreement really mean
good labellers?
6The human factor IIevaluation of annotations
WP3
kappa etc.
engineering
past WP9
7and later on
- mapping of labels onto cover classes
- sparse data
- classification performance
- embedding into the application task
- small number of alternatives
- criteria?
- dimensional labels adequate?
- human processing
- system restrictions cf. the story of 33 vs. 2
levels of accentuation
8The Sparse Data Problem
- un-balanced distribution (Pareto?)
- (too) few for robust training
- down- or up-sampling necessary for testing
- looking for "interesting" ("provocative"?)
data does this mean to beg the question?
9The Sparse Data Problem Some Frequencies,
word-based
not neutral in - /23
(27/9.6)
10.3/4.6
15.4/8 with/without emph.
- scenario-specific ironic vs. motherese/reprimandi
ng - emphatic in-between
- rare birds in AIBO surprised, helpless, bored
consensus labelling
10Towards New Dimensions
- from categories to dimensions
- confusion matrices similarities
?Non-Metrical Multi-Dimensional Solution (NMDS)
1111 emotional user state labels,
data-driven,word-based
- joyful
- surprised
- motherese
- neutral (default)
- rest (wast-paper-basket, non-neutral)
- bored
- helpless, hesitant
- emphatic (possibly indicating problems)
- touchy (irritated)
- angry
- reprimanding
effort 10-15 times real-time
12confusion matrix majority voting 3/5 vs. rest
if 2/2/1, both 2/2 as maj. vot. ("pre-emphasis")
A T R J M E N W S B H
Angry 43 13 12 00 00 12 18 00 00 00 00
Touchy 04 42 11 00 00 13 23 00 00 02 00 Reprim.
03 15 45 00 01 14 18 00 00 00 00 Joyful 00 00 01
54 02 07 32 00 00 00 00 Mother. 00 00 01 00 61 04
30 00 00 00 00 Emph. 01 05 06 00 01 53 29 00 00
00 00 Neutral 00 02 01 00 02 13 77 00 00 00
00 Wast-p. 00 07 06 00 08 19 21 32 00 01
01 Surpr. 00 00 00 00 00 20 40 00 40 00 00 Bored
00 14 01 00 01 12 28 01 00 39 00 Helpl. 00 01
00 02 00 12 37 03 00 00 41
R reprimanding, W waste paper basket category
13"traditional" emotional dimensions in
feeltraceVALENCE and AROUSAL
14NMDS 2-dimensional solution with 7
labels,relative majority with pre-empasis
orientation ?
valence ?
15and back
- from categories to dimensions
- what about the way back?
- automatic clustering?
- thresholds
- ....
16Towards New Quality Measures
- ?
- Stefan Steidl
- Entropy-Based Evaluation of Decoders
17Handling of Databases
- http//www.phonetik.uni-muenchen.de/Forschung/BITS
/index.html - Publications
- The Production of Speech Corpora (ISBN
3-8330-0700-1) - The Validation of Speech Corpora (ISBN
3-8330-0700-1) - The Production of Speech Corpora
- Florian Schiel, Christoph Draxler
- Angela Baumann, Tania Ellbogen, Alexander Steffen
- Version 2.5 June 1, 2004
18CEICES
- Combining Efforts for Improving automatic
Classification of Emotional user States, a
"forced co-operation" initiative under the
guidance of HUMAINE - evaluation of annotations
- assessment of F0 extraction algorithms
- assessment of impact of single feature (classes)
- improvement of classification performance via
sharing of features
19Ingredients of CEICES
- speech data German AIBO database
- annotations
- functional, emotional user states, word-based
- (prosodic peculiarities, word-based)
- manually corrected
- segment boundaries for words
- F0
- specifications of Train/Vali/Test, etc.
- reduction of effort ASCII file sharing via
portal - forced co-operation via agreement
20corr. F0
aut. F0
wordbound.
pitch-lab.
21Agreement
- open for non-HUMAINE partners
- nominal fee for distribution and handling
- commitments
- to share labels and extracted feature values
- to use specified sub-samples
- expected outcome
- assessment of F0 extraction, impact of features,
... - set of feature classes/vectors with evaluation
- common publication(s)
22some statements
- annotation has to be data-driven
- there is no bad labellers
- classification results have to be used for
labelling assessment - automatic labelling is not good enough - or,
maybe you should call it extraction - each label type has to be mapped onto very few
categorical classes at the end of the day
23Thank you for your attention