Title: Prosody in Spoken Language Understanding
1Prosody in Spoken Language Understanding
- Gina Anne Levow
- University of Chicago
- January 4, 2008
- NLP Winter School 2008
2- U Give me the price for ATT.
- U Give me the price for ATT.
- U Give me the price for ATT.
- U Give me the price for American Telephone and
Telegraph.
3Error Correction Spiral
- U Give me the price for ATT.
- S Hewlett Packard was 83 ¾, up 2 ½ since
yesterday. - U Give me the price for ATT.
- S Data General was at 10 ½ up a quarter.
- U Give me the price for ATT.
- S Hewlett Packard was 83 ¾, up 2 ½.
- U Give me the price for American Telephone and
Telegraph. - S Excuse me?
4Roadmap
- Corrections A motivating example
- Defining prosody
- Why prosody?
- Challenges in prosody
- Prosody in language understanding
- Recognizing tone and pitch accent
- Spoken corrections, Topic segmentation
- Conclusions
5Defining Prosody
- Prosody
- Phonetic phenomena in speech than span more than
a single segment-suprasegmental - Prosody includes
- Stress, focus, tone, intonation, length/pause,
rhythm - Prosodic features include
- Pitch perceptual correlate of fundamental
frequency - f0 rate of vocal fold vibration
- Loudness/intensity, duration, segment quality
6Why Prosody?
- Prosody plays a crucial role
- At all levels of language
- Lexical, syntactic, pragmatic/discourse
- Establishes meaning
- Disambiguates sense and structure
- Across languages families
- Common physiological, articulatory basis
- In synthesis and recognition of fluent speech
7Prosody and the Lexicon
- Lexical Determines word identity
- Prosodic effect at the syllable level (minimal
unit) - Lexical stress syllable prominence
- Combination of length, pitch movement, loudness
- REcord (N) vs reCORD (V)
- Pitch accent can differentiate words in some
languages - Lexical tone tone languages, e.g. Chinese,
Punjabi - Pitch height (register) and/or shape (contour)
Ma (high) mother Ma (rising) hemp Ma (low)
horse Ma (falling) scold
8Prosody and Syntax
- Prosody can disambiguate structure
- Associated with chunking and attachment
- Not identical with syntactic phrase boundaries
- Prosody is predictable from syntax, except when
it isnt - Prosodic phrasing indicated by
- Some combination of pause, change in pitch
9Chunking, or phrasing
- A1 I met Mary and Elenas mother at the mall
yesterday. - A2 I met Mary and Elenas mother at the mall
yesterday.
Example from Jennifer Venidetti
10Punctuation Prosody Humor
- A panda goes into a restaurant and has a meal.
Just before he leaves he takes out a gun and
fires it. The irate restaurant owner says Why
did you do that? The panda replies, I'm a
panda. Look it up.The restaurateur goes to his
dictionary and under panda finds black and
white arboreal, bear like creatures eats, shoots
and leaves.
11Prosody in Pragmatics Discourse
- Focus
- Prominence, new information pitch accent
- October eleventh
- Sentence type, dialogue act
- Statement vs. declarative question Its
raining (?) - Discourse Structure (Topic), Emotion
from Shih, Prosody Learning and Generation
12Challenges in Prosody I
- Highly variable
- Actual realization differs from ideal
- Speaker variation
- Gender, vocal track differences, idiosyncrasy
- Tonal coarticulation
- Neighboring tones influence (like segmental)
- Underlying fall can become rise
- Parallel encoding
- Effects at multiple levels realized
simultaneously
13 Challenges in Prosody II
- Challenges for learning
- Lack of training data
- Sparseness
- Many prosodic phenomena are infrequent
- E.g., non-declarative utterances, topic
boundaries, contrastive accents, etc - Challenging for machine learning methods
- Costs of labeling
- Many prosodic events require expert labeling
- Need large corpus to attest
- Time-consuming, expensive
14- Context and Learning in Multilingual Tone and
Pitch Accent Recognition
15Strategy Context
- Common model across languages
- Pure acoustic-prosodic model
- No word label, POS, lexical stress info
- English, Mandarin Chinese (also Cantonese,
isiZulu) - Exploit contextual information
- Features from adjacent syllables, phrase contour
- Analyze impact of
- Context position, context encoding, context type
- gt 12.5 reduction in error over no context
16Data Collections
- English (Ostendorf et al, 95)
- Boston University Radio News Corpus, f2b
- Manually annotated, aligned, syllabified
- 4 Pitch accent labels, aligned to syllables
- Mandarin
- TDT2 Voice of America Mandarin Broadcast News
- Automatically aligned, syllabified
- 4 main tones, neutral
17Local Feature Extraction
- Uniform representation for tone, pitch accent
- Motivated by Pitch Target Approximation Model
- Tone/pitch accent target exponentially approached
- Linear target height, slope (Xu et al, 99)
- Base features
- Pitch, Intensity max, mean, min, range
- (Praat, speaker normalized)
- Pitch at 5 points across voiced region
- Duration
- Initial, final in phrase
- Slope
- Linear fit to last half of pitch contour
18Context Features
- Local context
- Extended features
- Pitch max, mean, adjacent points of preceding,
following syllables - Difference features
- Difference between
- Pitch max, mean, mid, slope
- Intensity max, mean
- Of preceding, following and current syllable
- Phrasal context
- Compute collection average phrase slope
- Compute scalar pitch values, adjusted for slope
19Classification Experiments
- Classifier Support Vector Machine
- Linear kernel
- Multiclass formulation
- SVMlight (Joachims), LibSVM (Cheng Lin 01)
- 41 training / test splits
- Experiments Effects of
- Context position preceding, following, none,
both - Context encoding Extended/Difference
- Context type local, phrasal
20Results Local Context
Context Mandarin Tone English Pitch Accent
Full 74.5 81.3
Extend PrePost 74 80.7
Extend Pre 74 79.9
Extend Post 70.5 76.7
Diffs PrePost 75.5 80.7
Diffs Pre 76.5 79.5
Diffs Post 69 77.3
Both Pre 76.5 79.7
Both Post 71.5 77.6
No context 68.5 75.9
21Results Local Context
Context Mandarin Tone English Pitch Accent
Full 74.5 81.3
Extend PrePost 74 80.7
Extend Pre 74 79.9
Extend Post 70.5 76.7
Diffs PrePost 75.5 80.7
Diffs Pre 76.5 79.5
Diffs Post 69 77.3
Both Pre 76.5 79.7
Both Post 71.5 77.6
No context 68.5 75.9
22Results Local Context
Context Mandarin Tone English Pitch Accent
Full 74.5 81.3
Extend PrePost 74 80.7
Extend Pre 74 79.9
Extend Post 70.5 76.7
Diffs PrePost 75.5 80.7
Diffs Pre 76.5 79.5
Diffs Post 69 77.3
Both Pre 76.5 79.7
Both Post 71.5 77.6
No context 68.5 75.9
23Discussion Local Context
- Any context information improves over none
- Preceding context information consistently
improves over none or following context
information - English Generally more context features are
better - Mandarin Following context can degrade
- Little difference in encoding (Extend vs Diffs)
-
- Consistent with phonetic analysis (Xu) that
carryover coarticulation is greater than
anticipatory
24Results Discussion Phrasal Context
Phrase Context Mandarin Tone English Pitch Accent
Phrase 75.5 81.3
No Phrase 72 79.9
- Phrase contour compensation enhances recognition
- Simple strategy
- Use of non-linear slope compensate may improve
25Context Summary
- Employ common acoustic representation
- Tone (Mandarin), pitch accent (English)
- Cantonese 64 68 with RBF kernel
- SVM classifiers - linear kernel 76, 81
- Local context effects
- Up to gt 20 relative reduction in error
- Preceding context greatest contribution
- Carryover vs anticipatory
- Phrasal context effects
- Compensation for phrasal contour improves
recognition
26Strategy Training
- Challenge
- Can we use the underlying acoustic structure of
the language through unlabeled examples to
reduce the need for expensive labeled training
data? - Exploit semisupervised and unsupervised learning
- Semi-supervised Laplacian SVM
- K-means and asymmetric k-lines clustering
- Substantially outperform baselines
- Can approach supervised levels
27Data Collections Processing
- English (as before)
- Boston University Radio News Corpus, f2b
- Binary Unaccented vs accented
- 4-way Unaccented, High, Downstepped High, Low
- Mandarin
- Lab speech data (Xu, 1999)
- 5 syllable utterances vary tone, focus position
- In-focus, pre-focus, post-focus
- TDT2 Voice of America Mandarin Broadcast News
- 4-way High, Mid-rising, Low, High falling
- isiZulu (as before)
- Read web sentences
- 2-way High vs low
28Semi-supervised Learning
- Approach
- Employ small amount of labeled data
- Exploit information from additional presumably
more available unlabeled data - Few prior examples several weakly supervised
(Wong et al, 05) - Classifier
- Laplacian SVM (Sindhwani,BelkinNiyogi 05)
- Semi-supervised variant of SVM
- Exploits unlabeled examples
- RBF kernel, typically 6 nearest neighbors,
transductive
29Experiments
- Pitch accent recognition
- Binary classification Unaccented/Accented
- 1000 instances, proportionally sampled
- Labeled training 200 unacc, 100 acc
- 80 accuracy (cf. 84 w/15x labeled SVM)
- Mandarin tone recognition
- 4-way classification n(n-1)/2 binary classifiers
- 400 instances balanced 160 labeled
- Clean lab speech- in-focus-94
- cf. 99 w/SVM, 1000s train 85 w/SVM 160
training samples - Broadcast news 70
- Cf. lt 50 w/SVM 160 training samples
30Unsupervised Learning
- Question
- Can we identify the tone structure of a language
from the acoustic space without training? - Analogous to language acquisition
- Significant recent research in unsupervised
clustering - Established approaches k-means
- Spectral clustering (Shi Malik 97, Fischer
Poland 2004) asymmetric k-lines - Little research for tone
- Self-organizing maps (Gauthier et al,2005)
- Tones identified in lab speech using f0
velocities - Cluster-based bootstrapping (Narayanan et al,
2006) - Prominence clustering (Tambourini 05)
31Contrasting Clustering
- Contrasts
- Clustering 2-16 clusters, label w/most freq
class - 3 Spectral approaches
- Perform spectral decomposition of affinity matrix
- Asymmetric k-lines (Fischer Poland 2004)
- Symmetric k-lines (Fischer Poland 2004)
- Laplacian Eigenmaps (Belkin, Niyogi, Sindhwani
2004) - Binary weights, k-lines clustering
- K-means Standard Euclidean distance
- of clusters 2-16
- Best results gt 78
- 2 clusters asymmetric k-lines gt 2 clusters
kmeans - Larger clusters all similar
32Contrasting Learners
33Tone Clustering I
- Mandarin four tones
- 400 samples balanced
- 2-phase clustering 2-5 clusters each
- Asymmetric k-lines, k-means clustering
- Clean read speech
- In-focus syllables 87 (cf. 99 supervised)
- In-focus and pre-focus 77 (cf. 93 supervised)
- Broadcast news 57 (cf. 74 supervised)
- K-means requires more clusters to reach k-lines
level
34Tone Structure
First phase of clustering splits high/rising from
low/falling by slope Second phase by pitch height
35Conclusions
- Common prosodic framework for tone and pitch
accent recognition - Contextual modeling enhances recognition
- Local context and broad phrase contour
- Carryover coarticulation has larger effect for
Mandarin - Exploiting unlabeled examples for recognition
- Semi- and Un-supervised approaches
- Best cases approach supervised levels with less
training - Exploits acoustic structure of tone and accent
space
36Error Correction Spiral
- U Give me the price for ATT.
- S Hewlett Packard was 83 ¾, up 2 ½ since
yesterday. - U Give me the price for ATT.
- S Data General was at 10 ½ up a quarter.
- U Give me the price for ATT.
- S Hewlett Packard was 83 ¾, up 2 ½.
- U Give me the price for American Telephone and
Telegraph. - S Excuse me?
37Recognizing Spoken Corrections
- Spoken Corrections
- Recognize user attempts to correct ASR failures
- Compare original input to repeat corrections
- Significant differences
- Corrections increases in duration, pause
/length, final fall - Increases in pitch accent for misrecognitions
- Automatic recognition with decision trees,
boosting - Distinguish corrective/not (human level)
- Key features raw/normalized duration, pause
- Identify specific word being corrected
- Key features highest pitch, widest pitch range
38The ProblemSpeech Topic Segmentation
- Separate audio stream into component topics
On "World News Tonight" this Thursday, another
bad day on stock markets, all over the world
global economic anxiety. Another massacre in
Kosovo, the U.S. and its allies prepare to do
something about it. Very slowly. And the
millennium bug, Lubbock Texas prepares for
catastrophe, Bangalore, in India, sees only
profit.
39Is It Possible in Mandarin?
40Recognizing Shifts in Topic Turn
- Topic Turn boundaries in English Mandarin
- Initial syllables
- Significantly higher pitch, loudness than final
- Lexical and prosodic cues
- Cue words, tfidf similarity pitch, loudness,
silence - Automatic recognition with decision trees,
boosting - Voting to combine text, prosody, silence 97
accuracy - Key features
- Pause pitch, loudness contrast between syllables
41Conclusions Opportunities
- Prosody
- Rich source of information for languages
- Challenging due to variation, paucity of data
- Can be successfully employed, with learning, to
improve language understanding - Pitch accent, tone, dialogue act, turn, topic,
- Unrestricted conversational, multi-party,
multimodal speech much more challenging - Increased variability, interaction with
non-verbal evidence
42Thanks
- Dinoj Surendran, Siwei Wang, Yi Xu
- V. Sindhwani, M. Belkin, P. Niyogi I. Fischer
J. Poland T. Joachims C-C. Cheng C. Lin - This work supported by NSF Grant 0414919
- http//people.cs.uchicago.edu/levow/tai
43Phrasing can disambiguate
Mary Elenas mother
mall
I met Mary and Elenas mother at the mall
yesterday
One intonation phrase with relatively flat
overall pitch range.
44Phrasing can disambiguate
Elenas mother
mall
Mary
I met Mary and Elenas mother at the mall
yesterday
Separate phrases, with expanded pitch movements.
45Lists of numbers, nouns
- twenty.eight.five
- ninety.four.three
- seventy.three.seven
- forty.seven.seven
- seventy.seven.seven
- coffee cake and cream
- chocolate ice cream and cake
- fish fingers and bottles
- cheese sandwiches and milk
- cream buns and chocolate
from Prosody on the Web tutorial on chunking
46Clustering
- Pitch accent clustering
- 4 way distinction 1000 samples, proportional
- 2-16 clusters constructed
- Assign most frequent class label to each cluster
- Classifier
- Asymmetric k-lines
- context-dependent kernel radii, non-spherical
- gt 78 accuracy
- 2 clusters asymmetric k-lines best
- Context effects
- Vector w/preceding context vs vector with no
context comparable