Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2410.17457

mmWave-Whisper: Phone Call Eavesdropping and Transcription Using Millimeter-Wave Radar

Suryoday Basak·Abhijeeth Padarthi·Mahanth Gowda
View original source ↗Read source document ↗Export metadata ↓

Abstract

This paper introduces mmWave-Whisper, a system that demonstrates the feasibility of full-corpus automated speech recognition (ASR) on phone calls eavesdropped remotely using off-the-shelf frequency modulated continuous wave (FMCW) millimeter-wave radars. Operating in the 77-81 GHz range, mmWave-Whisper captures earpiece vibrations from smartphones, converts them into audio, and processes the audio to produce speech transcriptions automatically. Unlike previous work that focused on loudspeakers or limited vocabulary, this is the first work to perform such a speech recognition by handling large vocabulary and full sentences on earpiece vibrations from smartphones. This approach expands the potential of radar-audio eavesdropping. mmWave-Whisper addresses challenges such as the lack of large scale training datasets, low SNR, and limited frequency information in radar data through a systematic pipeline designed to leverage synthetic training data, domain adaptation, and inference by incorporating OpenAI's Whisper automatic speech recognition model. The system achieves a word accuracy rate of 44.74% and a character accuracy rate of 62.52% over a range of 25 cm to 125 cm. The paper highlights emerging misuse modalities of AI as the technology evolves rapidly.

Explore related subjects

cs.SDeess.AS

Keep this discovery

BibTeXRIS

Suryoday Basak, Abhijeeth Padarthi, Mahanth Gowda. 2024-10-22. mmWave-Whisper: Phone Call Eavesdropping and Transcription Using Millimeter-Wave Radar. https://arxiv.org/abs/2410.17457

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Xiaomi-CocktailASR-1 Technical Report

Recently, large language model (LLM) based ASR models have achieved significant progress, yet they generally lack support for multi-speaker scenarios, where the cocktail party problem remains a critical bottleneck for further advancing ASR. Existing TS-ASR methods, including end-to-end architectures with speaker embeddings and latest LLM-based explorations suffer from degraded single-speaker performance and the inability to reject when the target speaker is absent. In this paper, we propose Xiaomi-CocktailASR-1, an LLM-based end-to-end TS-ASR architecture. By utilizing reference speech as voiceprint prompts, it directly transcribes the target speaker's speech without requiring speech separation. Xiaomi-CocktailASR-1 maintains competitive performance in single-speaker scenarios, comparable to mainstream ASR models. It also features a negative sample rejection capability, outputting empty text when the target speaker is absent from the mixed speech. Additionally, Xiaomi-CocktailASR-1 supports a Chain-of-Thought (CoT) reasoning mode to provide explicit reasoning steps. Extensive experiments on various synthetic and real-world multispeaker benchmarks demonstrate that Xiaomi-CocktailASR-1 achieves state-of-the-art performance, effectively addressing the cocktail party problem through a unified architecture that balances multispeaker and single-speaker recognition accuracy, along with rejection capability.

cs.SD↗
arXiv2026

EConv-TasNet: Efficient Conv-TasNet for Effective Speech Separation

Conv-TasNet has served as a strong baseline for time-domain speech separation, and many studies have extended it with advanced architectures such as dual-path networks, U-Nets, and attention mechanisms. However, these methods often introduce high computational cost and complexity, limiting their deployment in resource-constrained scenarios. To address this issue, we propose eConv-TasNet, an efficient variant of Conv-TasNet that improves both effectiveness and efficiency without relying on resource-intensive modules. The proposed model consists of a group-wise early-splitting (GES) module and a multi-group feature aggregation (MGFA) module. GES generates discriminative speaker embeddings at intermediate stages, while MGFA progressively aggregates these group-level representations for refined mask estimation. Experimental results show that eConv-TasNet reduces model size by 22.4%, accelerates inference by 18.9%, and improves SI-SNRi by 14.0%-28.0% across three public benchmarks. Moreover, it achieves competitive performance compared with state-of-the-art methods while requiring significantly fewer parameters and lower inference cost. These results demonstrate a favorable efficiency-effectiveness trade-off for edge deployment.

cs.SD↗