cs 6998 computational approach to emotional speech instructor: prof. julia hirschberg

14
CS 6998 Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg Columbia University 12/21/2009 Julia’s Little Helper: A Real-time Demo of Cantonese/Mandarin Emotional Speech Detection William Y. Wang Computer Science Suzanne Yuen Mechanical Engineering

Upload: abner

Post on 21-Jan-2016

31 views

Category:

Documents


0 download

DESCRIPTION

Julia’s Little Helper : A Real-time Demo of Cantonese/Mandarin Emotional Speech Detection. Suzanne Yuen Mechanical Engineering. William Y. Wang Computer Science. CS 6998 Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg Columbia University 12/21/2009. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

CS 6998 Computational Approach to Emotional Speech

Instructor: Prof. Julia HirschbergColumbia University 12/21/2009

Julia’s Little Helper:

A Real-time Demo of Cantonese/Mandarin

Emotional Speech Detection

William Y. WangComputer Science

Suzanne YuenMechanical Engineering

Page 2: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Review

1. Target Languages: Cantonese (9 tones) , Mandarin (4 tones)

2. Target Emotions: Anger and Gladness

3. Lexical Features: ASR using a HMM acoustic model trained on

Mandarin Broadcast News [1] and a simple hand-written decoding

dictionary.

4. Prosodic Features: Energy and Tonal Features

5. Real-time drawing of pitch contour, waveform and energy.

6. A text-to-speech agent to greet and teach user how to use this

demo.

[1] Yang Shao, Lan Wang, E-Seminar: an Audio-guide e-Learning System, IEEE International Workshop

on Education Technology and Training (ETT) 2008.

Page 3: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Lexical Scoring 1-3pts Energy 1 pt Tone 1 pt

Page 4: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Dictionary of Affects in Language

by Dr. Cynthia Whissell

Words Pleasantness

Activeness

Imagery

affect 1.7500 1.8571 1.6

affection

2.7778 2.2500 2.0

success 2.8571 1.8000 1.4

successes

3.0000 2.0000 1.4Total words: 8742 words were included. Source: It was actually developed using various sources, for example, college student essays, interviews and teenagers description of their own emotion state. So, it can have a broad coverage and avoid biased data.

Page 5: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Sentence Lexical Scoring

Words Pleasantness

I 2.3750

won 2.5556

Best 2.5455

Paper 1.2857

Award 2.8333

“I won best paper award!”

Score = (2.375 + 2.5556 + 2.5455 + 1.2857 + 2.8333) / 5 = 2.319

Page 6: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Machine Translation

English Word

Pleasantness Chinese Translation

alcohol 1.7143 酒精 (alcohol)

alcoholic 1.0000 酒精 (alcohol)

muscles 2.3333 肌肉 (muscles)

muscular 2.6250 肌肉 (muscles)

Multilingual Challenges: English Chinese

Page 7: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Encoding and Mapping

Mandarin Word

Pinyin(Phone Set)

Corresponding Cantonese

PinyinMapping

喜欢 (like) xi3 hua1n 喜欢 (like) hei2 fu1n

睡觉 (sleep) shui4 jiao4 训觉 (sleep) fen4 gao4

Tasks:1. Mandarin Pinyin (Phone set used by Acoustic Model)2. Mandarin Cantonese

Note that not all words in Mandarin have theirs’ exact and direct mappings in Cantonese words and vice versa.

3. Cantonese Pinyin

Page 8: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Text-to-speech Engine

1. Implement the text-to-speech engine.

2. “Play with” a text-to-speech engine.

3. Engine: TruVoice

Lernout & Hauspie Speech Products, or L&H

Went bankrupt in 2001

technology now owned by Nuance

Page 9: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

L&H TTS Functionality

• Developed in 1997

• Advanced text pre-processing and no vocabulary restrictions

• User-definable pronunciation dictionary

• Accurately pronounces surnames and place names

• Flexible pitch, volume and speech rate

• Intonation support for punctuation

Page 10: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Test Overview

1. Participants –

1. gender: 6 male, 6 female

2. Native Language – 6 Mandarin, 6 Cantonese

2. Two Parts

1. JLH module and self-rating (24 lines total)

2. Perception test – Rating lines from others (72)

Page 11: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Sentences

1. Three types – questions, exclamations, statements

2. Randomized order of sentences for each participant

3. Examples:

In Lexicon Not In Lexicon

Angry & Not Angry

Insane!  痴 (chi1) 线 (xian4) !

You're wrong. 你 (ni3) 错 (cuo4) 了(le,liao3,liao4) 。

Glad & Not Glad When's the wedding?  什 (shen2,shi2) 么(ma,me) 时 (shi2) 候(hou4) 结 (jie1,jie2) 婚 (hun1) ?

It's bubbly?  有 (you3,you4) 泡(pao1,pao4) 泡(pao1,pao4) 吗(ma2,ma3,ma) ?

Page 12: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Analysis

1. Plan to examine differences and affects of following:

1. Ratings - JLH star rating, self rating, & 3 perception ratings

2. Language – Cantonese, Mandarin

3. Gender – female, male

4. Sentence structure – exclamation, question, & statement

2. Interesting points –

1. Huge range of Chinese accents

2. Tones of words may change depending on previous words

(such as English a mug vs. an umbrella)

3. Variations in colloquial speech, addressed by using Chinese

script

Page 13: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Future Work

1. Improve the prosodic analysis. More features should be explored.

2. Improve the lexical scoring. Use POS tagger or other NLP tools to

weigh different constituents of recognized sentence.

3. Finer-grain the emotion types and investigate the differences.

4. Study translational divergence in English-Chinese MT .

Page 14: CS 6998  Computational Approach to Emotional Speech Instructor: Prof. Julia Hirschberg

Demo