![](\"https:\/\/media.wired.com\/photos\/62969e5ac54d2f0c8603ffab\/master\/pass\/security-voice-ai.jpg\"\/)
<\/p>\n<\/p>\n
Credit to Author: Matt Burgess| Date: Wed, 01 Jun 2022 11:00:00 +0000<\/strong><\/p>\n Matt Burgess<\/a><\/span><\/span><\/p>\n To revist this article, visit My Profile, then View saved stories<\/a>.<\/p>\n To revist this article, visit My Profile, then View saved stories<\/a>.<\/p>\n Your voice reveals<\/span> more about you than you realize. To the human ear, your voice can instantly give away your mood, for example\u2014it\u2019s easy to tell if you\u2019re excited or upset. But machines can learn a lot more: inferring your age, gender, ethnicity, socio-economic status, health conditions, and beyond. Researchers have even been able to generate images of faces<\/a> based on the information contained in individuals\u2019 voice data.<\/p>\n As machines become better at understanding you through your voice, companies are cashing in. Voice recognition systems\u2014from Siri and Alexa to those using your voice as your password\u2014have proliferated in recent years as artificial intelligence and machine learning have unlocked the ability to understand not just what you are saying but who you are. Big Voice may be a $20 billion<\/a> industry within a few years. And as the market grows, privacy-focused researchers are increasingly searching for ways to protect people from having their voice data used against them.<\/p>\n Both the words you say and how you say them can be used to identify you, says Emmanuel Vincent, a senior research scientist specializing in voice technologies at France\u2019s National Institute for Research in Digital Science and Technology (Inria), but this is only the beginning. \u201cYou will also find other pieces of information about your emotions or your medical condition,\u201d Vincent says.<\/p>\n \u201cThese additional pieces of information help build a more complete profile\u2014then this would be used for all sorts of targeted advertisements,\u201d Vincent says. As well as your voice data potentially feeding into the vast realm of data used to show you online ads, there\u2019s also the risk that hackers could access the location where your voice data is stored and use it to impersonate you. A small number of these cloning incidents have already happened<\/a>, proving the value your voice holds. Simple robocall scams have also recorded people saying \u201cyes\u201d to use the confirmation in payment scams<\/a>.<\/p>\n Last year, TikTok changed its privacy policies and started collecting the voiceprints<\/a>\u2014a loose term for the data your voice contains\u2014of people in the US alongside other biometric data, such as your faceprint. More broadly, call centers are using AI to analyze people\u2019s \u201cbehavior and emotion\u201d<\/a> during phone calls and evaluate the \u201ctone, pace, and pitch of every single word\u201d to develop<\/a> profiles of people and increase sales<\/a>. \u201cWe\u2019re almost in a situation where the systems to recognize who you are and link everything together exist, but the protection is not there\u2014and it\u2019s still quite far away from being readily usable,\u201d says Henry Turner, who researched the security of voice systems at the University of Oxford.<\/p>\n Your voice is produced through a complex process<\/a> involving the lungs and your voice box, throat, nose, mouth, and sinuses. More than a hundred muscles are activated when you speak, says R\u00e9becca Kleinberger, a voice researcher at the MIT Media Lab. \u201cIt's also very much the brain,\u201d Kleinberger says.\u00a0<\/p>\n Researchers are experimenting with four ways to enhance privacy for your voice, says Natalia Tomashenko, a researcher at Avignon University, France, who has been studying voice and is the first author<\/a> of a research paper on the results of a voice privacy engineering challenge<\/a>. None of the methods are perfect, but they are being explored as possible ways to boost privacy in the infrastructure processing your voice data.<\/p>\n First is obfuscation, which tries to completely hide who the speaker is. Think of a Hollywood depiction of a hacker totally distorting their voice over a phone call as they explain a devilish plot or ransom (or hacktivist collective Anonymous\u2019s promotional videos<\/a>). Simple voice-changing hardware<\/a> allows anyone to quickly change the sound of their voice. More advanced speech-to-text-to-speech systems can transcribe what you\u2019re saying and then reverse the process<\/a> and say it in a new voice<\/a>.<\/p>\n Second, Tomashenko says, researchers are looking at distributed and federated learning<\/a>\u2014where your data doesn\u2019t leave your device but machine learning models still learn to recognize speech by sharing their training with a bigger system<\/a>. Another approach involves building encrypted infrastructure to protect people\u2019s voices from snooping. However, most efforts are focused on voice anonymization.<\/p>\n Anonymization attempts to keep your voice sounding human while stripping out as much of the information that could be used to identify you as possible. Speech anonymization efforts currently involve two separate strands: anonymizing the content of what someone is saying by deleting or replacing any sensitive words in files before they are saved and anonymizing the voice itself. Most voice anonymization efforts at the moment involve passing someone\u2019s voice through experimental software that will change some of the parameters in the voice signal to make it sound different. This can involve altering the pitch, replacing segments of speech with information from other voices, and synthesizing the final output.<\/p>\n Does anonymization technology work? Male and female voice clips<\/a> that were anonymized as part of the Voice Privacy Challenge in 2020 definitely do sound different. They\u2019re more robotic, sound slightly pained and could\u2014to some listeners at least\u2014be from a different person than the original voice clips. \u201cI think it can already guarantee a much higher level of protection than doing nothing, which is the current status,\u201d says Vincent, who has been able to reduce how easy it is to identify people in anonymization research<\/a>. However, humans aren\u2019t the only listeners. Rita Singh, an associate professor in Carnegie Mellon University\u2019s Language Technologies Institute, says that total de-identification of the voice signal is not possible, as machines will always have the potential to make links between attributes and individuals, even connections that aren\u2019t clear to humans. \u201cIs the anonymization with respect to a human listener or is it with respect to a machine listener?\u201d says Shri Narayanan, a professor of electrical and computer engineering at the University of Southern California.<\/p>\n \u201cTrue anonymization is not possible without completely changing the voice,\u201d Singh says. \u201cWhen you completely change the voice, then it's not the same voice.\u201d Despite this, it is still worth developing voice-privacy technology, Singh adds, as no privacy or security system is totally secure. Fingerprints and face identification systems on iPhones have been spoofed<\/a> in the past<\/a>, but overall, they\u2019re still an effective method of protecting people\u2019s privacy.<\/p>\n Your voice is increasingly being used as a way to verify your identity. For example, a growing number of banks and other companies are analyzing your voiceprints, with your permission, to replace your password. There\u2019s also the potential for voice analysis to detect illness before other signs are obvious. But the technology to clone or fake someone\u2019s voice is advancing quickly.<\/p>\n If you have a few minutes of someone\u2019s voice recorded, or in some instances a few seconds, it\u2019s possible to recreate that voice using machine learning\u2014The Simpsons\u2019<\/em> voice actors could be replaced by deep fake voice clones<\/a>, for instance. And commercial tools for recreating voices are readily available online<\/a>. \u201cThere\u2019s definitely more work in speaker identification and producing speech to text and text to speech than there is in protecting people from any of those technologies,\u201d Turner says.<\/p>\n Many of the voice anonymization techniques being developed at the moment are still a long way from being used in the real world. When they are ready to be used it\u2019s likely that companies will have to implement tools themselves, to protect their customers' privacy\u2014there\u2019s currently little individuals can do to protect their own voice. Avoiding calls with call centers or companies that use voice analysis, and not using voice assistants, could limit how much your voice is recorded and reduce possible attack opportunities.<\/p>\n But the biggest protections may come from legal cases and protections. Europe\u2019s GDPR<\/a> covers biometric data, including people\u2019s voices, in its privacy protections. Guidelines say<\/a> people should be told how their data is being used and provide consent if they\u2019re being identified, and that some restrictions should be placed on personalization. Meanwhile, in the US, courts in Illinois\u2014 home to some of the strongest biometric laws in the country\u2014are increasingly inspecting cases involving people\u2019s voice data. McDonald\u2019s, Amazon, and Google are all facing judicial scrutiny<\/a> over how they use people\u2019s voice data. The decisions in these cases could lay down new rules for the protection of people\u2019s voices.<\/p>\n