SlideShare ist ein Scribd-Unternehmen logo
1 von 11
Aspiring Minds
www.aspiringminds.com
Spoken English Evaluation
Machine Learning with Crowd Intelligence
Varun Aggarwal
Presented at KDD, 2015, ACL 2015
Problem Statement & Motivation
Importance of spoken English
English language has a very high socio-economic impact – with people speaking the language fluently
reported to earn 30-50% more than their peers who don’t.
Grading spoken English in a scalable way needed by companies, training organization and also
individuals.
Problem Statement
Scalable grading of spontaneous English speech, as good as experts.
Why are automated methods not accurate?
Speaker independent Speech
recognition for spontaneous
speech is a hard problem!
Proposed system architecture
Crowdsourcing helps us get
accurate transcriptions. Crowd grades
also help
improve!Crowd Grades
FA Features
Crowdsourcing task
Crowdsourcing task
Worker quality control
• Each worker is assigned a risk level which reflects the
quality of his past work.
• Based on the state, number and when to give a gold
standard task is determined.
Supervised learning setup
Experiment Details
• Sample Size : 566
• 319 India
• 247 from Philippines
Expert Grading
• Two expert raters
• Overall score based on Pronunciation/Fluency
Content-Org/Grammar.
• Inter-rater correlation ~0.8.
The learning task
• Modelling done separately for Indian and Philippines
set.
• Linear ridge regression, Neural Networks and SVM
regression with different kernels were used to build
the models.
Case study
• Studied deployment of proposed algorithm in
Philippines.
• Event had 500 applicants for the role of a
customer support executive. The scoring
algorithm was tested on a subset of 150 students.
• Internal expert graded each candidate’s speech as
hirable or not-hireable.
Features used
We use three classes of features
• Force Alignment features (FA) and
• The speech sample is forced aligned on the crowdsourced transcription.
• Features like– rate of speech, position and length of pauses, log likelihood of recognition, posterior probability,
hesitations and repetitions, etc are derived.
• Natural Language Processing features (NLP).
• Surface level features : number of words, complexity or difficulty of words and the number of common words
used.
• Semantic features like the coherency in text, context of the words spoken, sentiment of the text and grammar
correctness.
• Crowd Grades (CG)
• Crowd provides scores on - pronunciation, fluency, content organization and grammar.
• These grades are combined to form a composite score.
Experiment and Results
Crowdsourced transcriptions + Crowd grades outperforms all other methods
Accuracy nears inter expert agreement (~0.8).
Summing it up
• Svar provides an automated assessment of candidate’s pronunciation and fluency.
• Crowdsourcing, in addition to NLP feature, renders reliable composite scores.
• Speech assessments can be made scalable with accuracy nearly matching experts’ opinion.

Weitere ähnliche Inhalte

Was ist angesagt?

The best ways to study english
The best ways to study englishThe best ways to study english
The best ways to study englishcorneliacalin1
 
Ielts speaking
Ielts speakingIelts speaking
Ielts speakingxuandoc
 
Test effort estimation
Test effort estimationTest effort estimation
Test effort estimationramesh kumar
 
API Test Automation Tips and Tricks
API Test Automation Tips and TricksAPI Test Automation Tips and Tricks
API Test Automation Tips and Trickstesthive
 
Test Automation Frameworks: Assumptions, Concepts & Tools
Test Automation Frameworks: Assumptions, Concepts & ToolsTest Automation Frameworks: Assumptions, Concepts & Tools
Test Automation Frameworks: Assumptions, Concepts & ToolsAmit Rawat
 
Soa testing soap ui (2)
Soa testing   soap ui (2)Soa testing   soap ui (2)
Soa testing soap ui (2)Knoldus Inc.
 
Four wheel steering system
Four wheel steering systemFour wheel steering system
Four wheel steering systemThakur Singh
 
Electric vehicles (1)
Electric vehicles (1)Electric vehicles (1)
Electric vehicles (1)WasimAbdulla1
 
How to Prepare for a Behavioral Interview
How to Prepare for a Behavioral InterviewHow to Prepare for a Behavioral Interview
How to Prepare for a Behavioral InterviewCyndi McCabe
 
Forming English Questions
Forming English QuestionsForming English Questions
Forming English QuestionsLIA Tangerang
 
Engine Management System/ ECU
Engine Management System/ ECUEngine Management System/ ECU
Engine Management System/ ECUSahil Mohile
 
What is Web Testing?
What is Web Testing?   What is Web Testing?
What is Web Testing? QA InfoTech
 
Automatic Emergency Braking
Automatic Emergency BrakingAutomatic Emergency Braking
Automatic Emergency Brakingmani kanta
 
IELTS Writing Task 1 Process Diagrams
IELTS Writing Task 1 Process DiagramsIELTS Writing Task 1 Process Diagrams
IELTS Writing Task 1 Process DiagramsDavid Wills
 
How to Design a Successful Test Automation Strategy
How to Design a Successful Test Automation Strategy How to Design a Successful Test Automation Strategy
How to Design a Successful Test Automation Strategy Impetus Technologies
 
learning-spoken-english
learning-spoken-englishlearning-spoken-english
learning-spoken-englishPepe Pita
 

Was ist angesagt? (20)

The best ways to study english
The best ways to study englishThe best ways to study english
The best ways to study english
 
PET guidelines
PET guidelinesPET guidelines
PET guidelines
 
Unit 7 powerpoint
Unit 7 powerpointUnit 7 powerpoint
Unit 7 powerpoint
 
Ielts speaking
Ielts speakingIelts speaking
Ielts speaking
 
Test effort estimation
Test effort estimationTest effort estimation
Test effort estimation
 
API Test Automation Tips and Tricks
API Test Automation Tips and TricksAPI Test Automation Tips and Tricks
API Test Automation Tips and Tricks
 
Test Automation Frameworks: Assumptions, Concepts & Tools
Test Automation Frameworks: Assumptions, Concepts & ToolsTest Automation Frameworks: Assumptions, Concepts & Tools
Test Automation Frameworks: Assumptions, Concepts & Tools
 
Soa testing soap ui (2)
Soa testing   soap ui (2)Soa testing   soap ui (2)
Soa testing soap ui (2)
 
Four wheel steering system
Four wheel steering systemFour wheel steering system
Four wheel steering system
 
Electric vehicles (1)
Electric vehicles (1)Electric vehicles (1)
Electric vehicles (1)
 
How to Prepare for a Behavioral Interview
How to Prepare for a Behavioral InterviewHow to Prepare for a Behavioral Interview
How to Prepare for a Behavioral Interview
 
Forming English Questions
Forming English QuestionsForming English Questions
Forming English Questions
 
Engine Management System/ ECU
Engine Management System/ ECUEngine Management System/ ECU
Engine Management System/ ECU
 
What is Web Testing?
What is Web Testing?   What is Web Testing?
What is Web Testing?
 
Component testing with cypress
Component testing with cypressComponent testing with cypress
Component testing with cypress
 
Automatic Emergency Braking
Automatic Emergency BrakingAutomatic Emergency Braking
Automatic Emergency Braking
 
IELTS Writing Task 1 Process Diagrams
IELTS Writing Task 1 Process DiagramsIELTS Writing Task 1 Process Diagrams
IELTS Writing Task 1 Process Diagrams
 
Job interview
Job interview Job interview
Job interview
 
How to Design a Successful Test Automation Strategy
How to Design a Successful Test Automation Strategy How to Design a Successful Test Automation Strategy
How to Design a Successful Test Automation Strategy
 
learning-spoken-english
learning-spoken-englishlearning-spoken-english
learning-spoken-english
 

Andere mochten auch

Aspiring Minds | Automata
Aspiring Minds | Automata Aspiring Minds | Automata
Aspiring Minds | Automata Aspiring Minds
 
16720032294774_Sirallapu_Anitha_corpReport
16720032294774_Sirallapu_Anitha_corpReport16720032294774_Sirallapu_Anitha_corpReport
16720032294774_Sirallapu_Anitha_corpReportanitha sirallapu
 
Aspiring Minds | AM Situations
Aspiring Minds | AM SituationsAspiring Minds | AM Situations
Aspiring Minds | AM SituationsAspiring Minds
 
Aspiring Minds | Outcomes using test scores
Aspiring Minds | Outcomes using test scoresAspiring Minds | Outcomes using test scores
Aspiring Minds | Outcomes using test scoresAspiring Minds
 
Prediction of Salary From Profiles
Prediction of Salary From ProfilesPrediction of Salary From Profiles
Prediction of Salary From ProfilesSohom Ghosh
 
Campus Performace Report
Campus Performace ReportCampus Performace Report
Campus Performace ReportSayed Ali
 
Am cat workshop part 1
Am cat workshop part 1Am cat workshop part 1
Am cat workshop part 1vanatteveldt
 
Recruitments & assessment industry
Recruitments & assessment industryRecruitments & assessment industry
Recruitments & assessment industryRahul Koul
 
Campus New Proposal.
Campus New Proposal.Campus New Proposal.
Campus New Proposal.Sayed Ali
 
Aspiring Minds | Labor market insights
Aspiring Minds | Labor market insightsAspiring Minds | Labor market insights
Aspiring Minds | Labor market insightsAspiring Minds
 
About Youth4work - Integrated Talent Solutions
About Youth4work - Integrated Talent SolutionsAbout Youth4work - Integrated Talent Solutions
About Youth4work - Integrated Talent SolutionsYouth4work.com
 
Youth4work Marketing & Advertising Solutions
Youth4work Marketing & Advertising SolutionsYouth4work Marketing & Advertising Solutions
Youth4work Marketing & Advertising SolutionsYouth4work.com
 
Institute Performance Solutions
Institute Performance Solutions Institute Performance Solutions
Institute Performance Solutions Youth4work.com
 
Campus Hiring Made Easy
Campus Hiring Made Easy Campus Hiring Made Easy
Campus Hiring Made Easy Youth4work.com
 
Humanika Consulting presentation 2012b english
Humanika Consulting presentation 2012b englishHumanika Consulting presentation 2012b english
Humanika Consulting presentation 2012b englishSeta Wicaksana
 

Andere mochten auch (18)

Aspiring Minds | Automata
Aspiring Minds | Automata Aspiring Minds | Automata
Aspiring Minds | Automata
 
16720032294774_Sirallapu_Anitha_corpReport
16720032294774_Sirallapu_Anitha_corpReport16720032294774_Sirallapu_Anitha_corpReport
16720032294774_Sirallapu_Anitha_corpReport
 
Aspiring Minds | AM Situations
Aspiring Minds | AM SituationsAspiring Minds | AM Situations
Aspiring Minds | AM Situations
 
Aspiring Minds | Outcomes using test scores
Aspiring Minds | Outcomes using test scoresAspiring Minds | Outcomes using test scores
Aspiring Minds | Outcomes using test scores
 
Amcat Certificate
Amcat CertificateAmcat Certificate
Amcat Certificate
 
Prediction of Salary From Profiles
Prediction of Salary From ProfilesPrediction of Salary From Profiles
Prediction of Salary From Profiles
 
Campus Performace Report
Campus Performace ReportCampus Performace Report
Campus Performace Report
 
Am cat workshop part 1
Am cat workshop part 1Am cat workshop part 1
Am cat workshop part 1
 
Recruitments & assessment industry
Recruitments & assessment industryRecruitments & assessment industry
Recruitments & assessment industry
 
Campus New Proposal.
Campus New Proposal.Campus New Proposal.
Campus New Proposal.
 
Aspiring Minds | Labor market insights
Aspiring Minds | Labor market insightsAspiring Minds | Labor market insights
Aspiring Minds | Labor market insights
 
About Youth4work - Integrated Talent Solutions
About Youth4work - Integrated Talent SolutionsAbout Youth4work - Integrated Talent Solutions
About Youth4work - Integrated Talent Solutions
 
Youth4work Marketing & Advertising Solutions
Youth4work Marketing & Advertising SolutionsYouth4work Marketing & Advertising Solutions
Youth4work Marketing & Advertising Solutions
 
Institute Performance Solutions
Institute Performance Solutions Institute Performance Solutions
Institute Performance Solutions
 
Amcat & nac
Amcat & nac Amcat & nac
Amcat & nac
 
Campus Hiring Made Easy
Campus Hiring Made Easy Campus Hiring Made Easy
Campus Hiring Made Easy
 
Humanika Consulting presentation 2012b english
Humanika Consulting presentation 2012b englishHumanika Consulting presentation 2012b english
Humanika Consulting presentation 2012b english
 
Amcat 2
Amcat 2Amcat 2
Amcat 2
 

Ähnlich wie Aspiring Minds | Svar

Elements of language learning - an analysis of how different elements of lang...
Elements of language learning - an analysis of how different elements of lang...Elements of language learning - an analysis of how different elements of lang...
Elements of language learning - an analysis of how different elements of lang...PrithaVashisht1
 
How to Asess Forieng Language Proficiency #wlclassroom
How to Asess Forieng Language Proficiency #wlclassroomHow to Asess Forieng Language Proficiency #wlclassroom
How to Asess Forieng Language Proficiency #wlclassroomJoshua Cabral
 
actfl-opi-reliability-2012
actfl-opi-reliability-2012actfl-opi-reliability-2012
actfl-opi-reliability-2012Hyder Abadin
 
Reliability of an oral test (1)
Reliability of an oral test (1)Reliability of an oral test (1)
Reliability of an oral test (1)huma nasir
 
Language Testing
Language Testing Language Testing
Language Testing edac4co
 
Relating language examinations to the common European reference levels of lan...
Relating language examinations to the common European reference levels of lan...Relating language examinations to the common European reference levels of lan...
Relating language examinations to the common European reference levels of lan...Nelly Zafeiriades
 
Assessment in Writing.pptx
Assessment in Writing.pptxAssessment in Writing.pptx
Assessment in Writing.pptxRifkaFaidah
 
Summary of all the chapters
Summary of all the chaptersSummary of all the chapters
Summary of all the chapterskashmasardar
 
Classroom Language Test
Classroom Language TestClassroom Language Test
Classroom Language TestKin Susansi
 
Language Assessment - Standardized Testing by EFL Learners
Language Assessment - Standardized Testing by EFL LearnersLanguage Assessment - Standardized Testing by EFL Learners
Language Assessment - Standardized Testing by EFL LearnersEFL Learning
 
introducing language testing and assessment
 introducing language testing  and assessment introducing language testing  and assessment
introducing language testing and assessmentNajah M. Algolaip
 
Language Assessment : Kinds of tests and testing
Language Assessment : Kinds of tests and testingLanguage Assessment : Kinds of tests and testing
Language Assessment : Kinds of tests and testingMusfera Nara Vadia
 
Language testing final
Language testing finalLanguage testing final
Language testing finaledac4co
 
Wiseman Facets of L2 writing ability
Wiseman Facets of L2 writing abilityWiseman Facets of L2 writing ability
Wiseman Facets of L2 writing abilityCynthia Wiseman
 
Wiseman facets of l2 writing tesol 2012
Wiseman facets of l2 writing tesol 2012Wiseman facets of l2 writing tesol 2012
Wiseman facets of l2 writing tesol 2012Cynthia Wiseman
 

Ähnlich wie Aspiring Minds | Svar (20)

Clear Talk Brochure
Clear Talk BrochureClear Talk Brochure
Clear Talk Brochure
 
Elements of language learning - an analysis of how different elements of lang...
Elements of language learning - an analysis of how different elements of lang...Elements of language learning - an analysis of how different elements of lang...
Elements of language learning - an analysis of how different elements of lang...
 
How to Asess Forieng Language Proficiency #wlclassroom
How to Asess Forieng Language Proficiency #wlclassroomHow to Asess Forieng Language Proficiency #wlclassroom
How to Asess Forieng Language Proficiency #wlclassroom
 
actfl-opi-reliability-2012
actfl-opi-reliability-2012actfl-opi-reliability-2012
actfl-opi-reliability-2012
 
Reliability of an oral test (1)
Reliability of an oral test (1)Reliability of an oral test (1)
Reliability of an oral test (1)
 
Language Testing
Language Testing Language Testing
Language Testing
 
Relating language examinations to the common European reference levels of lan...
Relating language examinations to the common European reference levels of lan...Relating language examinations to the common European reference levels of lan...
Relating language examinations to the common European reference levels of lan...
 
Assessment in Writing.pptx
Assessment in Writing.pptxAssessment in Writing.pptx
Assessment in Writing.pptx
 
Summary of all the chapters
Summary of all the chaptersSummary of all the chapters
Summary of all the chapters
 
Classroom Language Test
Classroom Language TestClassroom Language Test
Classroom Language Test
 
Assessment purposes and approaches
Assessment purposes and approachesAssessment purposes and approaches
Assessment purposes and approaches
 
Language Assessment - Standardized Testing by EFL Learners
Language Assessment - Standardized Testing by EFL LearnersLanguage Assessment - Standardized Testing by EFL Learners
Language Assessment - Standardized Testing by EFL Learners
 
introducing language testing and assessment
 introducing language testing  and assessment introducing language testing  and assessment
introducing language testing and assessment
 
Language Assessment : Kinds of tests and testing
Language Assessment : Kinds of tests and testingLanguage Assessment : Kinds of tests and testing
Language Assessment : Kinds of tests and testing
 
HR Management
HR Management   HR Management
HR Management
 
Language testing final
Language testing finalLanguage testing final
Language testing final
 
iTEP
iTEPiTEP
iTEP
 
Phyllis Anderson SHL
Phyllis Anderson SHLPhyllis Anderson SHL
Phyllis Anderson SHL
 
Wiseman Facets of L2 writing ability
Wiseman Facets of L2 writing abilityWiseman Facets of L2 writing ability
Wiseman Facets of L2 writing ability
 
Wiseman facets of l2 writing tesol 2012
Wiseman facets of l2 writing tesol 2012Wiseman facets of l2 writing tesol 2012
Wiseman facets of l2 writing tesol 2012
 

Kürzlich hochgeladen

Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Manik S Magar
 
Scanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsScanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsRizwan Syed
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubKalema Edgar
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxNavinnSomaal
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr BaganFwdays
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxhariprasad279825
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLScyllaDB
 
WordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your BrandWordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your Brandgvaughan
 
Powerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time ClashPowerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time Clashcharlottematthew16
 
TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024Lonnie McRorey
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostZilliz
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 3652toLead Limited
 
How to write a Business Continuity Plan
How to write a Business Continuity PlanHow to write a Business Continuity Plan
How to write a Business Continuity PlanDatabarracks
 
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxMerck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxLoriGlavin3
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfAlex Barbosa Coqueiro
 
Connect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck PresentationConnect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck PresentationSlibray Presentation
 
Advanced Computer Architecture – An Introduction
Advanced Computer Architecture – An IntroductionAdvanced Computer Architecture – An Introduction
Advanced Computer Architecture – An IntroductionDilum Bandara
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...Fwdays
 

Kürzlich hochgeladen (20)

Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!Anypoint Exchange: It’s Not Just a Repo!
Anypoint Exchange: It’s Not Just a Repo!
 
Scanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL CertsScanning the Internet for External Cloud Exposures via SSL Certs
Scanning the Internet for External Cloud Exposures via SSL Certs
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding Club
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptx
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptx
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQL
 
WordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your BrandWordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your Brand
 
Powerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time ClashPowerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time Clash
 
TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024
 
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptxE-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
E-Vehicle_Hacking_by_Parul Sharma_null_owasp.pptx
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365
 
How to write a Business Continuity Plan
How to write a Business Continuity PlanHow to write a Business Continuity Plan
How to write a Business Continuity Plan
 
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxMerck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdf
 
Connect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck PresentationConnect Wave/ connectwave Pitch Deck Presentation
Connect Wave/ connectwave Pitch Deck Presentation
 
Advanced Computer Architecture – An Introduction
Advanced Computer Architecture – An IntroductionAdvanced Computer Architecture – An Introduction
Advanced Computer Architecture – An Introduction
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
 

Aspiring Minds | Svar

  • 1. Aspiring Minds www.aspiringminds.com Spoken English Evaluation Machine Learning with Crowd Intelligence Varun Aggarwal Presented at KDD, 2015, ACL 2015
  • 2. Problem Statement & Motivation Importance of spoken English English language has a very high socio-economic impact – with people speaking the language fluently reported to earn 30-50% more than their peers who don’t. Grading spoken English in a scalable way needed by companies, training organization and also individuals. Problem Statement Scalable grading of spontaneous English speech, as good as experts.
  • 3. Why are automated methods not accurate? Speaker independent Speech recognition for spontaneous speech is a hard problem!
  • 4. Proposed system architecture Crowdsourcing helps us get accurate transcriptions. Crowd grades also help improve!Crowd Grades FA Features
  • 6. Crowdsourcing task Worker quality control • Each worker is assigned a risk level which reflects the quality of his past work. • Based on the state, number and when to give a gold standard task is determined.
  • 7. Supervised learning setup Experiment Details • Sample Size : 566 • 319 India • 247 from Philippines Expert Grading • Two expert raters • Overall score based on Pronunciation/Fluency Content-Org/Grammar. • Inter-rater correlation ~0.8. The learning task • Modelling done separately for Indian and Philippines set. • Linear ridge regression, Neural Networks and SVM regression with different kernels were used to build the models.
  • 8. Case study • Studied deployment of proposed algorithm in Philippines. • Event had 500 applicants for the role of a customer support executive. The scoring algorithm was tested on a subset of 150 students. • Internal expert graded each candidate’s speech as hirable or not-hireable.
  • 9. Features used We use three classes of features • Force Alignment features (FA) and • The speech sample is forced aligned on the crowdsourced transcription. • Features like– rate of speech, position and length of pauses, log likelihood of recognition, posterior probability, hesitations and repetitions, etc are derived. • Natural Language Processing features (NLP). • Surface level features : number of words, complexity or difficulty of words and the number of common words used. • Semantic features like the coherency in text, context of the words spoken, sentiment of the text and grammar correctness. • Crowd Grades (CG) • Crowd provides scores on - pronunciation, fluency, content organization and grammar. • These grades are combined to form a composite score.
  • 10. Experiment and Results Crowdsourced transcriptions + Crowd grades outperforms all other methods Accuracy nears inter expert agreement (~0.8).
  • 11. Summing it up • Svar provides an automated assessment of candidate’s pronunciation and fluency. • Crowdsourcing, in addition to NLP feature, renders reliable composite scores. • Speech assessments can be made scalable with accuracy nearly matching experts’ opinion.

Hinweis der Redaktion

  1. - A high number of jobs in knowledge economies across the globe require English. - Companies want to be able to test scalable - Training insttns need to be scalably test and provide feedback
  2. great transcription to know what is spoken; once we know what is spoken- we can compare the pronunciation of the candidate with a good pronounciation of the word; But because transcription is bad; we don't get to know what is spoken; this makes feature derivation inaccurate Pure ML ~0.5
  3. So we use two sets of features: One derived from aligning speech sample crowd transcription AND the other directly crowd grades
  4. Easy usability This is where people transcribe This is where people grade
  5. We had a novel idea to give every turker a state: the state tells the current reliability of the worker; it depends on how many gold standards s/he did write; A high reliability worker sees less gold standards and low reliability sees more of them. This helps manage risk with money spent on gold standards.
  6.  SUPERVISED LEARNING; OUTPUT is expert grades; and we are trying to do with our system We use several techniques like NN, SVM, etc. with crossvalidation