按年被提到
20182018 · 3 次2026
视觉、语音与图像生成›语音与音频生成·arXiv:1801.01944 · cs.LG

Audio Adversarial Examples: Targeted Attacks on Speech-to-Text

Nicholas Carlini, David Wagner · 2018
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 4 人首次 2018-01-11 · 最近 2024-01-08
4不同的人
2活跃月份
4讨论串
1,087被引(OpenAlex)
arXiv 摘要(候选层不写导读)

We construct targeted audio adversarial examples on automatic speech recognition. Given any audio waveform, we can produce another that is over 99.9% similar, but transcribes as any phrase we choose (recognizing up to 50 characters per second of audio). We apply our white-box iterative optimization-based attack to Mozilla's implementation DeepSpeech end-to-end, and show it has a 100% success rate. The feasibility of this attack introduce a new domain to study adversarial examples.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论dijksterhuis · 2024-01

“Decent write up:+1: Recommend reading on the topic - Biggio & Rolia "Wild Patterns" review paper for the thorough security perspective (and historical accuracy cough): https://arxiv.org/pdf/1712.03141.pdf - Carlini & Wagner attack a.k.a. the gold standard of adversarial machine learning research papers: https://arxiv.org/abs/1608.04644 - Carlini & Wa…”

Attacks on machine learning models原评论 ↗
被提交为 story · 3 次