W <scp>hisper</scp> T <scp>est</scp> : A Voice-Control-based Library for iOS UI Automation
Bibliographic record
Abstract
Dynamic analysis and UI automation are essential for scalable detection of privacy leaks, vulnerabilities, and malicious code in mobile apps. While the Android ecosystem offers a variety of tools, options for iOS apps are limited and require either access to the app source code or jailbreaking the test device. To address this gap, we introduce WhisperTest, an open-source iOS UI automation library that operates without jailbreaking. WhisperTest is based on a newly designed approach that leverages Apple's Voice Control accessibility feature to interact with app or system UIs via text-to-speech. During interactions, WhisperTest monitors the device system logs in real time and scrapes the UI via screenshots and accessibility audits to recover app state changes. We demonstrate WhisperTest's capabilities through a diverse set of tasks, including a web privacy measurement and a fully-automated dynamic analysis of 200 child-directed iOS apps. To overcome the challenges of automating apps with diverse UI designs, WhisperTest optionally integrates multimodal large language models to reason about context and interact with system permission prompts, consent dialogs, subscription prompts, and age gates. Our exploratory analysis of children's apps uncovers widespread use of third-party tracking, limited recognition of user consent, and unencrypted HTTP requests. Overall, we show that WhisperTest enables scalable dynamic analysis of iOS applications across diverse tasks, contributing to a safer and more transparent mobile ecosystem.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.011 |
| Meta-epidemiology (narrow) | 0.003 | 0.002 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.004 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.057 | 0.052 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".