Targeted Audio Adversarial Examples
12 weeks ago by jm
This is phenomenal:
The audio examples on this page are impressive -- a little bit of background noise, such as you might hear on a telephone call with high compression, hard to perceive if you aren't listening out for it.
Paper here: https://arxiv.org/abs/1801.01944
(Via Parker Higgins, https://twitter.com/xor )
papers
audio
adversarial-classification
neural-networks
speech-to-text
speech
recognition
voice
attacks
exploits
via:xor
We have constructed targeted audio adversarial examples on speech-to-text transcription neural networks: given an arbitrary waveform, we can make a small perturbation that when added to the original waveform causes it to transcribe as any phrase we choose.
In prior work, we constructed hidden voice commands, audio that sounded like noise but transcribed to any phrases chosen by an adversary. With our new attack, we are able to improve this and make an arbitrary waveform transcribe as any target phrase.
The audio examples on this page are impressive -- a little bit of background noise, such as you might hear on a telephone call with high compression, hard to perceive if you aren't listening out for it.
Paper here: https://arxiv.org/abs/1801.01944
(Via Parker Higgins, https://twitter.com/xor )
12 weeks ago by jm
Computer says no: Irish vet fails oral English test needed to stay in Australia
This is idiotic. Computer-based voice recognition is in no way reliable enough for this kind of job. It's automated Kafkaesque bureaucracy -- "computer says no". Shame on Oz
(via James Kelleher)
via:etienneshrdlu
kafkaesque
bureaucracy
computer-says-no
voice-recognition
australia
immigration
english
voice
testing
august 2017 by jm
An Irish veterinarian with degrees in history and politics has been unable to convince a machine she can speak English well enough to stay in Australia.
Louise Kennedy is a native English speaker, has excellent grammar and a broad vocabulary. She holds two university degrees – both obtained in English – and has been working in Australia as an equine vet on a skilled worker visa for the past two years.
But she is now scrambling for other visa options after a computer-based English test – scored by a machine – essentially handed her a fail in terms of convincing immigration officers she can fluently speak her own language.
This is idiotic. Computer-based voice recognition is in no way reliable enough for this kind of job. It's automated Kafkaesque bureaucracy -- "computer says no". Shame on Oz
(via James Kelleher)
august 2017 by jm
AIY Projects Voice Kit
may 2017 by jm
This looks nifty!
(via sergio)
voice
google
aiy-projects
cardboard
hacks
raspberry-pi
This project demonstrates how to get a natural language recognizer up and running and connect it to the Google Assistant. Along with everything the Google Assistant already does, you can add your own question and answer pairs. All in a handy little cardboard cube, powered by a Raspberry Pi.
(via sergio)
may 2017 by jm
Pink Trombone
march 2017 by jm
A model of how voice sounds are produced. Pretty cool
voice
phonetics
sound
mouth
science
march 2017 by jm
A Guide to Communication, Shotcalling, and Etiquette in Competitive Overwatch
october 2016 by jm
Excellent post on team voice comms tactics. Many tips here
voice
voice-comms
gaming
overwatch
communication
strats
october 2016 by jm
Where do 'mama'/'papa' words come from?
sounds
voice
speech
babies
kids
phonetics
linguist
language
october 2015 by jm
The sounds came first — as experiments in vocalization — and parents adopted them as pet names for themselves.
If you open your mouth and make a sound, it will probably be an open vowel like /a/ unless you move your tongue or lips. The easiest consonants are perhaps the bilabials /m/, /p/, and /b/, requiring no movement of the tongue, followed by consonants made by raising the front of the tongue: /d/, /t/, and /n/. Add a dash of reduplication, and you get mama, papa, baba, dada, tata, nana.
That such words refer to people (typically parents or other guardians) is something we have imposed on the sounds and incorporated into our languages and cultures; the meanings don’t inhere in the sounds as uttered by babies, which are more likely calls for food or attention.
october 2015 by jm
related tags
adversarial-classification ⊕ aiy-projects ⊕ attacks ⊕ audio ⊕ australia ⊕ babies ⊕ bureaucracy ⊕ cardboard ⊕ communication ⊕ computer-says-no ⊕ english ⊕ exploits ⊕ gaming ⊕ google ⊕ hacks ⊕ immigration ⊕ kafkaesque ⊕ kids ⊕ language ⊕ linguist ⊕ mouth ⊕ neural-networks ⊕ overwatch ⊕ papers ⊕ phonetics ⊕ raspberry-pi ⊕ recognition ⊕ science ⊕ sound ⊕ sounds ⊕ speech ⊕ speech-to-text ⊕ strats ⊕ testing ⊕ via:etienneshrdlu ⊕ via:xor ⊕ voice ⊖ voice-comms ⊕ voice-recognition ⊕Copy this bookmark: