How to "defeat" Google Audio Recapcha?

Overview of unCapcha - small tool to "pass thru" audio capcha 
26 October 2017   2710

Across the Internet, hundreds of thousands of sites rely on Google's reCaptcha system for defense against bots (in fact, Devpost uses reCaptcha when creating a new account). After a Google research team demonstrated a near complete defeat of the text reCaptcha in 2012, the reCaptcha system evolved to rely on audio and image challenges, historically more difficult challenges for automated systems to solve. Google has continually iterated on its design, releasing a newer and more powerful version as recently as just this year. Successfully demonstrating a defeat of this captcha system spells significant vulnerability for hundreds of thousands of popular sites.

What is unCapcha?

UnCaptcha system has attack capabilities written for the audio captcha. Using browser automation software, we can interact with the target website and engage with the captcha, parsing out the necessary elements to begin the attack. We rely primarily on the audio captcha attack - by properly identifying spoken numbers, we can pass the reCaptcha programmatically and fool the site into thinking our bot is a human. Specifically, unCaptcha targets the popular site Reddit by going through the motions of creating a new user, although unCaptcha stops before creating the user to mitigate the impact on Reddit.

Background

Google's reCaptcha system uses an advanced risk analysis system to determine programmatically how likely a given user is to be a human or a bot. It takes into account your cookies (and by extension, your interaction with other Google services), the speed at which challenges are solved, mouse movements, and (obviously) how successfully you solve the given task. As the system gets increasingly suspicious, it delivers increasingly difficult challenges, and requires the user to solve more of them. Researchers have already identified minor weaknesses with the reCaptcha system - 9 days of legitimate (ish) interaction with Google's services is usually enough to lower the system's suspicion level significantly.

How it works

The format of the audio captcha is a varied-length series of numbers spaced out read aloud at varied speeds, pitches, and accents through background noise. To attack this captcha, the audio payload is identified on the page, downloaded, and automatically split by locations of speech.

From there, each number audio bit is uploaded to 6 different free, online audio transcription services (IBM, Google Cloud, Google Speech Recognition, Sphinx, Wit-AI, Bing Speech Recognition), and these results are collected. Developers ensemble the results from each of these to probabilistically enumerate the most likely string of numbers with a predetermined heuristic. These numbers are then organically typed into the captcha, and the captcha is completed. From testing, team has seen 92%+ accuracy in individual number identification, and 85%+ accuracy in defeating the audio captcha in its entirety.

Python News Digest 21 - 27.03

Fast subsets of large datasets with Pandas, how to use Pandas read_html, new pip resolver release announced and more
27 March 2020   1031

Greetings! I hope your week went great! Here's new frontend technologies news digest.

Learn how to do a binary search in Python, check the introduction to SQLAlchemy ORM for Django developers, read about the Python code metrics and other cool things, prerpared for you.

Guides

  • Fast subsets of large datasets with Pandas and SQLite

Tutorial about fast subsets of large datasets with Pandas and SQLite

  • The Python math Module: Everything You Need to Know

This guide will teach you all about Python’s math module for higher-level mathematical functions

  • How to use Pandas read_html to Scrape Data from HTML Tables

Detailed tutorial on how to use Pandas read_html method for scraping data from HTML

  • More on service layers in Django

Another chapter of post about the drawbacks of using a service layer architecture for building Django apps.

Articles

  • New pip resolver to roll out this year

Developers says there will be big changes to how pip determinates what to install, based on the package requirements

Updates

  • Python 3.9.0a5

Early developer preview of Python 3.9 is now available

  • superstring.py

Library for heavy-text manipulation in Python

  • WhyNot

A Python sandbox for decision making in dynamics

  • PySimpleGUI-COVID19

Python written tools to help analyze the spread of the COVID-19 virus

  • Z3

Python-written theorem prover from Microsoft Research