audiocaptcha

23
AN IMPROVED AUDIO Submitted By :- Swapnil Singh 0816513057 I.T. - IIIrd year

Upload: swati-shukla

Post on 04-Jul-2015

28 views

Category:

Documents


5 download

TRANSCRIPT

Page 1: Audiocaptcha

AN IMPROVED AUDIO

Submitted By :-

Swapnil Singh

0816513057

I.T. - IIIrd year

Page 2: Audiocaptcha

CONTENTS

What are captchas? Problem with current audio captchas. Testing of current captchas. Categories of audio Captcha. Algorithm used and its details. Need for audio reCaptcha. Applications. Pitfalls. Conclusion.

2

Page 3: Audiocaptcha

3

WHAT ARE CAPTCHAS?

CAPTCHAs are tests generated by computers and generally passable by humans but not current computer programs.

Page 4: Audiocaptcha

4

THE PROBLEM WITH CURRENT AUDIO CAPTCHAS

In some cases the human passing rate is only 70%!

To make the CAPTCHAs secure, noise was injected into the audio files making it harder for both computers and humans to pass.

Page 5: Audiocaptcha

5

ARE CURRENT AUDIO CAPTCHAS SECURE?

A CAPTCHA is considered broken once a program can pass it 5% of the time.

Since the current audio CAPTCHAs use a limited vocabulary, it was possible for us to collect enough data to train a system that could pass the current audio CAPTCHAs more than 45% of the time.

Page 6: Audiocaptcha

6

HOW DID WE TEST THE CURRENT AUDIO CAPTCHAs?

Selected three different types of audio CAPTCHAs: google, reCAPTCHA, and digg

Collected 1000 CAPTCHAs per type of audio CAPTCHA to use for training and testing

Created an ASR system using machine learning techniques

Page 7: Audiocaptcha

THREE CATEGORIES OF AUDIO CAPTCHA reCAPTCHA audio captcha - multiple voices,

digits and background noise that is backwards speech

Google audio captcha- digits, single voice, backwards speech

Digg audio captcha- digits and letters, static/water for noise

7

Page 8: Audiocaptcha

8

THE ALGORITHM

Given the .wav file of an audio CAPTCHA Segmentation - selecting portions of the audio

which most likely are digits/letters Recognition

Extract features from the segmentClassify segment as digit/letter or noise and

output the label Stop once a maximum number of segments are

classified

Page 9: Audiocaptcha

9

ALGORITHM DETAILS - SEGMENTATION

CAPTCHAs were manually labeled and segmented. We created training segments using this information.

For testing, we chose the highest energy peaks in the test CAPTCHA and selected fixed size segments roughly centered at the peaks.

QuickTime™ and a decompressor

are needed to see this picture.

Page 10: Audiocaptcha

10

ALGORITHM DETAILS - FEATURES

We used three popular techniques for extracting features from speech to derive 5 sets of features from the audio.Mel-frequency cepstral coefficients (MFCC)Perceptual linear prediction (PLP)Relative spectral transform with PLP (RASTA-PLP)

Page 11: Audiocaptcha

11

ALGORITHM DETAILS - AdaBoost

Used decision stumps for weak classifiers For each type of audio CAPTCHA we created

enough classifiers to label a segment as a digit, letter, or noise.

Created 11 to 37 classifiers Each classifier returns a value which represents

its confidence that the segment should be labeled as digit letter or noise.

Page 12: Audiocaptcha

12

ALGORITHM DETAILS - SVM

Created a single multiclass classifier using all the training segments (from 900 CAPTCHAs)

Page 13: Audiocaptcha

13

ALGORITHM DETAILS - k-NN

Created 5 classifiers corresponding to each of the feature sets

Page 14: Audiocaptcha

14

THE ALGORITHM

Input: Audio CAPTCHA as an audio file Segmentation

Find the highest energy peak, and extract a fixed size segment centered at that peak

RecognitionExtract features from segmentGive segment to classifier and obtain label

Stop extracting segments once all segments have been labeled or a max solution size is reached.

Page 15: Audiocaptcha

15

Using three machine learning techniques to perform ASR on the CAPTCHAsAdaBoostSupport Vector

Machines (SVM)k-Nearest Neighbor

(k-NN)

0

10

20

30

40

50

60

70

80

%

GooglereCAPTCHA Digg

Exact Match Rate

AdaBoostSVMk-NN

ANALYSIS OF CURRENT AUDIO CAPTCHAs

Page 16: Audiocaptcha

16

THE GOAL

Make a secure audio CAPTCHA which will be easier for a human to pass and harder for a computer to pass.

Equate solving a CAPTCHA with doing some useful work. In other words, create an audio reCAPTCHA.

Page 17: Audiocaptcha

17

WHAT IS reCAPTCHA?

reCAPTCHA helps digitize text on which OCR fails by using the text as its CAPTCHA.

Since millions of people solve CAPTCHAs each day, millions of words get digitized each day!

Page 18: Audiocaptcha

18

Page 19: Audiocaptcha

19

THE AUDIO RECAPTCHA

Takes advantage of the human ability to understand words through context.

Will help transcribe digital audio on which ASR systems fail.

The audio being used was originally recorded with the intention that it should be easily understood by humans.

Page 20: Audiocaptcha

APPLICATIONS

Preventing Comment Spam in Blogs.  Protecting Website Registration.  Protecting Email Addresses From

Scrapers. Online Polls Preventing Dictionary Attacks. Worms and Spam.

20

Page 21: Audiocaptcha

21

ANALYSIS OF SECURITY

Speaker independent recognition is difficult. Open vocabularies make it even more difficult

for ASR systems AM broadcasts and .mp3 compression cause the

loss of important data needed for automatic analysis

Page 22: Audiocaptcha

22

CONCLUSION

CAPTCHAs need to be more accessible, yet remain secure and not too difficult for humans.

Deploy audio reCAPTCHA through reCAPTCHA site.

Help make knowledge captured in audio available in text form

Page 23: Audiocaptcha

Thank you

23