1 00:00:01,320 --> 00:00:07,240 Speaker 1: Welcome to Prognosis. I'm Laura Carlson. It's day three since 2 00:00:07,280 --> 00:00:13,319 Speaker 1: coronavirus was declared a global pandemic. Today's main story. New 3 00:00:13,360 --> 00:00:18,919 Speaker 1: technology promises it can spot early COVID nineteen symptoms simply 4 00:00:18,960 --> 00:00:23,760 Speaker 1: by listening to the sound of someone's voice. But first, 5 00:00:24,640 --> 00:00:36,519 Speaker 1: here's what happened in virus moons today. President Joe Biden's 6 00:00:36,560 --> 00:00:40,440 Speaker 1: pushed for another large COVID nineteen relief package got some 7 00:00:40,520 --> 00:00:45,800 Speaker 1: help today. A surprisingly weak January employment report helped make 8 00:00:45,840 --> 00:00:49,440 Speaker 1: the case for substantial aid, and the Senate took action 9 00:00:49,880 --> 00:00:54,240 Speaker 1: to smooth the legislative path for his proposal. Private sector 10 00:00:54,280 --> 00:00:58,840 Speaker 1: payrolls barely grew in January, while the restaurant and lodging 11 00:00:58,880 --> 00:01:03,640 Speaker 1: sector log almost six hundred thousand in job losses over 12 00:01:03,680 --> 00:01:07,319 Speaker 1: the past two months. That's according to the monthly jobs 13 00:01:07,319 --> 00:01:11,720 Speaker 1: report from the Labor Department out today. While the jobless 14 00:01:11,800 --> 00:01:15,120 Speaker 1: rate dipped to six point three percent, that was partly 15 00:01:15,280 --> 00:01:20,520 Speaker 1: because some Americans gave up looking for work. The United 16 00:01:20,640 --> 00:01:24,120 Speaker 1: Kingdom said it plans to offer everyone over the age 17 00:01:24,120 --> 00:01:28,480 Speaker 1: of fifty a first dose of coronavirus vaccine by May. 18 00:01:28,920 --> 00:01:32,479 Speaker 1: It's the first time that Boris Johnson's government has set 19 00:01:32,480 --> 00:01:37,279 Speaker 1: a target for the next stage of its vaccination program. 20 00:01:37,280 --> 00:01:40,440 Speaker 1: The plan was included in an announcement that also said 21 00:01:40,720 --> 00:01:44,479 Speaker 1: local and mayoral elections will go ahead on May six, 22 00:01:45,200 --> 00:01:49,960 Speaker 1: after they were delayed by a year due to the pandemic. Finally, 23 00:01:50,600 --> 00:01:54,320 Speaker 1: a study has found a surprisingly strong indicator of whether 24 00:01:54,440 --> 00:01:58,080 Speaker 1: someone infected with COVID nineteen has a higher risk of 25 00:01:58,160 --> 00:02:02,880 Speaker 1: severe disease and death. That indicator is whether they've had 26 00:02:03,080 --> 00:02:08,760 Speaker 1: pneumonia in the past. According to researchers at Harvard University, 27 00:02:09,240 --> 00:02:12,720 Speaker 1: a prior episode of pneumonia was the second greatest overall 28 00:02:12,840 --> 00:02:16,720 Speaker 1: risk factor for death from COVID nineteen. The study showed. 29 00:02:17,120 --> 00:02:21,360 Speaker 1: The researchers pointed out that by itself, a single pneumonia 30 00:02:21,440 --> 00:02:25,240 Speaker 1: case probably doesn't put someone at high risk. It's more 31 00:02:25,360 --> 00:02:29,840 Speaker 1: likely to be an indicator of underlying chronic disease that's 32 00:02:29,880 --> 00:02:39,120 Speaker 1: gone undiagnosed. And now for today's main story. When we're 33 00:02:39,160 --> 00:02:41,960 Speaker 1: coming down with a cold, or are feeling a bit stressed, 34 00:02:42,360 --> 00:02:45,880 Speaker 1: or perhaps even exhibiting the first symptoms of COVID nineteen, 35 00:02:46,760 --> 00:02:50,079 Speaker 1: minute changes to our voice are often one of the 36 00:02:50,120 --> 00:02:55,240 Speaker 1: first indicators that something is wrong. These vocal bio markers 37 00:02:55,320 --> 00:02:58,600 Speaker 1: are often beyond what a human can detect, But what 38 00:02:58,680 --> 00:03:03,760 Speaker 1: if an app on your phone could. Health reporter Michelle 39 00:03:03,760 --> 00:03:08,440 Speaker 1: fack Cortez recently spoke to David Luke, CEO of sound Health, 40 00:03:08,880 --> 00:03:11,640 Speaker 1: which is released an app that uses a person's voice 41 00:03:11,840 --> 00:03:17,200 Speaker 1: to detect early symptoms of respiratory illnesses, including COVID nineteen. 42 00:03:18,120 --> 00:03:21,680 Speaker 1: I asked Michelle what vocal bio markers can tell us. 43 00:03:22,880 --> 00:03:26,040 Speaker 1: You recently had a chance to speak with the CEO 44 00:03:26,280 --> 00:03:30,400 Speaker 1: of sound Health, which is released an app called sold one. 45 00:03:30,520 --> 00:03:32,560 Speaker 1: I was wondering if you could tell us a bit 46 00:03:32,600 --> 00:03:36,600 Speaker 1: about this app and how it relates to COVID nineteen. 47 00:03:38,080 --> 00:03:41,520 Speaker 1: This app is looking for vocal bio markers indications that 48 00:03:41,560 --> 00:03:44,960 Speaker 1: you might be actively infected with coronavirus and that the 49 00:03:45,040 --> 00:03:49,240 Speaker 1: illness could be having an impact on your voice. It 50 00:03:49,440 --> 00:03:52,920 Speaker 1: is something that is relatively easy to use. It's something 51 00:03:52,960 --> 00:03:57,920 Speaker 1: that's being adopted by companies and schools and other outlets 52 00:03:57,960 --> 00:04:00,320 Speaker 1: as a way to get an early heads up on 53 00:04:00,440 --> 00:04:03,880 Speaker 1: potential risk. David Leeu is the CEO of sound Health, 54 00:04:03,960 --> 00:04:06,240 Speaker 1: and he laid out how it works for us. Song 55 00:04:06,360 --> 00:04:10,320 Speaker 1: One is a app based product that can be downloaded 56 00:04:10,360 --> 00:04:13,800 Speaker 1: from any smartphone with that that app, you're able to 57 00:04:13,880 --> 00:04:16,400 Speaker 1: record six seconds of voice and we can give you 58 00:04:16,560 --> 00:04:21,359 Speaker 1: then a reading on your risk for having symptoms of 59 00:04:21,400 --> 00:04:25,160 Speaker 1: respiratory disease like COVID nineteen. In terms of how this 60 00:04:25,279 --> 00:04:29,680 Speaker 1: app can actually detect whether or not you have COVID 61 00:04:29,760 --> 00:04:34,000 Speaker 1: nineteen or another respiratory illness, I mean, how exactly is 62 00:04:34,000 --> 00:04:37,240 Speaker 1: it doing that through your voice? Your voice is an 63 00:04:37,320 --> 00:04:40,800 Speaker 1: integral part of you, and you know from your own 64 00:04:40,800 --> 00:04:43,440 Speaker 1: personal experience listening to your loved ones and people that 65 00:04:43,480 --> 00:04:47,039 Speaker 1: you know well, sometimes you can tell if somebody is 66 00:04:47,120 --> 00:04:49,960 Speaker 1: not feeling great, if there might be something going on 67 00:04:50,080 --> 00:04:54,599 Speaker 1: with them in terms of illness or perhaps anxiety or stress. 68 00:04:55,200 --> 00:04:59,120 Speaker 1: We're talking about things like the patterns in the rhythms, 69 00:04:59,480 --> 00:05:04,320 Speaker 1: the will intonation, any kind of variation in that process 70 00:05:04,440 --> 00:05:07,280 Speaker 1: is what they're looking for. And as you do it 71 00:05:07,720 --> 00:05:10,760 Speaker 1: over and over again, the app can get better at 72 00:05:10,839 --> 00:05:15,240 Speaker 1: knowing your own individual patterns comparing that to yourself and 73 00:05:15,400 --> 00:05:18,240 Speaker 1: to other people. David laid out for us the way 74 00:05:18,240 --> 00:05:23,080 Speaker 1: that it works. Bio Markers are any any information that 75 00:05:23,200 --> 00:05:27,000 Speaker 1: is given off by the body. Could be your come 76 00:05:27,080 --> 00:05:31,120 Speaker 1: from blood, could be come from saliva, your your heart rate. Um. 77 00:05:31,240 --> 00:05:33,240 Speaker 1: These are signals that are coming from your body that 78 00:05:33,279 --> 00:05:36,440 Speaker 1: can be measured right and can be recorded, and so 79 00:05:36,800 --> 00:05:41,719 Speaker 1: vocal bio markers are field that has really begun to 80 00:05:41,880 --> 00:05:44,320 Speaker 1: grow and explode in the last five years, I would say, 81 00:05:44,360 --> 00:05:47,760 Speaker 1: so it's fairly new when used in the health context. 82 00:05:48,240 --> 00:05:52,279 Speaker 1: Our company son has has already launched vocal bio markers 83 00:05:52,320 --> 00:05:56,360 Speaker 1: for detection of mental health conditions such as depression and 84 00:05:56,440 --> 00:05:59,960 Speaker 1: now respiratory illness. There are other companies in the space 85 00:06:00,040 --> 00:06:03,320 Speaker 1: stead of also UH used vocal bio markers for other 86 00:06:03,360 --> 00:06:08,240 Speaker 1: health conditions. So what is the science or you know, 87 00:06:08,360 --> 00:06:13,360 Speaker 1: the research that underpins this app How have they constructed 88 00:06:13,400 --> 00:06:18,480 Speaker 1: away that something on your phone can detect symptoms of 89 00:06:18,520 --> 00:06:22,360 Speaker 1: an illness or even COVID nineteen. There has been some 90 00:06:22,680 --> 00:06:26,720 Speaker 1: research done on this in order to determine how effective 91 00:06:26,760 --> 00:06:29,800 Speaker 1: the approaches. But of course it's important to realize this 92 00:06:29,920 --> 00:06:32,800 Speaker 1: isn't something that's approved by the Food and Drug Administration, 93 00:06:33,200 --> 00:06:35,880 Speaker 1: and you're ensurre isn't going to pay for it, but 94 00:06:36,000 --> 00:06:39,320 Speaker 1: it is an early indicator of potential danger. You can 95 00:06:39,320 --> 00:06:43,279 Speaker 1: think of it kind of like a thermometer or your 96 00:06:43,360 --> 00:06:46,240 Speaker 1: mom putting her hand on your forehead to see if 97 00:06:46,279 --> 00:06:49,279 Speaker 1: there's something going on with a layer of science and 98 00:06:49,320 --> 00:06:53,760 Speaker 1: technology on top of that. Really drilling down into the details. 99 00:06:54,920 --> 00:06:56,520 Speaker 1: You need to have a lot of experience and a 100 00:06:56,520 --> 00:06:59,280 Speaker 1: lot of data to have been and to have studied 101 00:06:59,320 --> 00:07:03,599 Speaker 1: this data. Voice data that is health condition labeled, meaning 102 00:07:03,760 --> 00:07:06,520 Speaker 1: we have now over a million voice samples that we 103 00:07:06,560 --> 00:07:11,360 Speaker 1: have collected and studied must have confirmed labels stating that 104 00:07:11,440 --> 00:07:15,320 Speaker 1: this person had the disease or the the health condition, 105 00:07:15,800 --> 00:07:18,040 Speaker 1: and so they've been diagnosed with that condition, and then 106 00:07:18,080 --> 00:07:20,840 Speaker 1: we study their voice from there. We also study people's 107 00:07:20,880 --> 00:07:24,840 Speaker 1: voices who are completely healthy and we compare those those voices. 108 00:07:25,360 --> 00:07:29,240 Speaker 1: When you have that larger volume of data, you're then 109 00:07:29,320 --> 00:07:33,800 Speaker 1: able to look at different groups of people men, women, uh, 110 00:07:33,840 --> 00:07:37,400 Speaker 1: You're able to look at different ages of people, different devices. 111 00:07:37,480 --> 00:07:41,200 Speaker 1: So many things come into consideration when you're analyzing voice. 112 00:07:41,640 --> 00:07:44,000 Speaker 1: When we're looking at different health conditions, then we can 113 00:07:44,080 --> 00:07:47,560 Speaker 1: zero in, for example, in respiratory disease or even depression, 114 00:07:47,880 --> 00:07:50,960 Speaker 1: we can zero in on the vocal features that are 115 00:07:51,000 --> 00:07:55,960 Speaker 1: most sensitive to change when somebody displays a symptom of 116 00:07:56,000 --> 00:07:59,080 Speaker 1: that disease. And again we are not diagnosing for disease. 117 00:07:59,360 --> 00:08:04,000 Speaker 1: We are comply monitoring and detecting the change in your voice. 118 00:08:04,240 --> 00:08:06,800 Speaker 1: Vocal features that are not able to be picked up 119 00:08:06,840 --> 00:08:11,200 Speaker 1: by the human ear, such as prosody, such as tonality, 120 00:08:11,920 --> 00:08:14,960 Speaker 1: even breaks and speech um. There are so many of 121 00:08:15,000 --> 00:08:17,559 Speaker 1: these vocal features that we do examine and put into 122 00:08:17,600 --> 00:08:21,840 Speaker 1: our machine learning models that help us then predict based 123 00:08:21,920 --> 00:08:26,000 Speaker 1: upon the data, who might have who has a higher 124 00:08:26,120 --> 00:08:30,080 Speaker 1: probability of having these symptoms of disease versus people who 125 00:08:30,120 --> 00:08:33,839 Speaker 1: are completely fine. Just to be clear, this is something 126 00:08:33,880 --> 00:08:38,640 Speaker 1: that can only pick up symptoms when you're already manifesting 127 00:08:38,679 --> 00:08:42,480 Speaker 1: symptoms of COVID nineteen or another respiratory illness. This isn't 128 00:08:42,520 --> 00:08:45,880 Speaker 1: something that could pick up whether or not you have 129 00:08:45,960 --> 00:08:49,880 Speaker 1: COVID nineteen if you happen to be asymptomatic. It's not 130 00:08:50,120 --> 00:08:53,680 Speaker 1: a diagnostic. It doesn't tell you that you are definitively 131 00:08:53,800 --> 00:08:58,880 Speaker 1: infected with coronavirus. But it is another layer, another indicator 132 00:08:59,160 --> 00:09:03,160 Speaker 1: that could help well determine if they are at risk. 133 00:09:03,520 --> 00:09:07,560 Speaker 1: The company is not claiming that it is something that 134 00:09:07,600 --> 00:09:11,960 Speaker 1: should be used in order to determine who is safe 135 00:09:11,960 --> 00:09:15,360 Speaker 1: for high risk situations in terms of you know, visiting 136 00:09:15,440 --> 00:09:18,079 Speaker 1: friends and family, are going to a nursing home, flying, 137 00:09:18,160 --> 00:09:21,640 Speaker 1: that sort of a thing. But it is another piece 138 00:09:21,760 --> 00:09:26,160 Speaker 1: of the puzzle. You know, we've been focusing on COVID nineteen, 139 00:09:26,320 --> 00:09:30,240 Speaker 1: but what else can these types of apps like so 140 00:09:30,320 --> 00:09:35,560 Speaker 1: on one detect using these vocal bio markers. Previous research 141 00:09:35,600 --> 00:09:39,280 Speaker 1: has looked at vocal bio markers for indications of other 142 00:09:39,960 --> 00:09:46,040 Speaker 1: potential health conditions, most specifically mental health issues like anxiety 143 00:09:46,160 --> 00:09:49,600 Speaker 1: and depression. Those are areas where we do get a 144 00:09:49,640 --> 00:09:54,720 Speaker 1: lot of change that happens, that comes through your voice. 145 00:09:55,440 --> 00:09:59,800 Speaker 1: Sometimes looking intently at that particular piece can give you 146 00:09:59,840 --> 00:10:02,520 Speaker 1: in it into how you're feeling and where your health 147 00:10:02,600 --> 00:10:06,880 Speaker 1: is going. I'm wondering about some of the potential risks 148 00:10:07,040 --> 00:10:10,000 Speaker 1: or problems. I mean, being able to speak into your 149 00:10:10,040 --> 00:10:12,480 Speaker 1: phone and it tell you whether or not you have 150 00:10:12,720 --> 00:10:15,960 Speaker 1: symptoms of either COVID nineteen or just a respiratory illness 151 00:10:16,000 --> 00:10:21,199 Speaker 1: could be very useful, but what are the risks, say 152 00:10:21,240 --> 00:10:24,560 Speaker 1: I'm thinking of um in terms of either patient privacy 153 00:10:24,720 --> 00:10:31,320 Speaker 1: or just misdiagnosis. Privacy and misdiagnosis are actually critical issues, 154 00:10:31,520 --> 00:10:34,280 Speaker 1: especially in this period of time that we're in right now. 155 00:10:34,760 --> 00:10:40,240 Speaker 1: Everyone is talking about privacy across every platform as we're 156 00:10:40,280 --> 00:10:44,640 Speaker 1: all interacting with each other online, but there are absolutely 157 00:10:44,840 --> 00:10:49,440 Speaker 1: issues about where this information is going, especially because in 158 00:10:49,480 --> 00:10:54,120 Speaker 1: many cases it's employers or other organizations that are gathering 159 00:10:54,120 --> 00:10:59,600 Speaker 1: this information. There is room for things to go sideways. Now, Michelle, 160 00:10:59,679 --> 00:11:03,080 Speaker 1: you've actually have the opportunity to try out sound one 161 00:11:03,120 --> 00:11:07,440 Speaker 1: for yourself. Walk us through what the process is of 162 00:11:07,440 --> 00:11:11,600 Speaker 1: of using the app. So I downloaded the app this morning, 163 00:11:12,400 --> 00:11:15,920 Speaker 1: and I used a key that the company gave me 164 00:11:16,160 --> 00:11:20,200 Speaker 1: full disclosure, and I answered a couple of questions including 165 00:11:20,200 --> 00:11:24,640 Speaker 1: my email address, my gender, my birthdate, and now I'm 166 00:11:24,679 --> 00:11:27,120 Speaker 1: ready to give it a shot. All right, walk us through, 167 00:11:27,640 --> 00:11:30,240 Speaker 1: you know, submitting a vocal sample. Here we go. So 168 00:11:30,320 --> 00:11:33,600 Speaker 1: I'm looking at a screen. It says sound one tap 169 00:11:33,679 --> 00:11:39,400 Speaker 1: here to get started inside concentric white circles. It's loading. 170 00:11:39,880 --> 00:11:42,959 Speaker 1: You have a questionnaire and a voice activity to complete. 171 00:11:43,400 --> 00:11:45,360 Speaker 1: My voice does feel a little scratchy today, so we'll 172 00:11:45,360 --> 00:11:48,880 Speaker 1: see how that goes. I definitely feel healthy, though, and 173 00:11:48,920 --> 00:11:51,000 Speaker 1: I have been pretty much in my basement for the 174 00:11:51,080 --> 00:11:53,200 Speaker 1: last month, so I don't think I'm high risk. Here 175 00:11:54,080 --> 00:11:55,960 Speaker 1: in this activity, I want you to take a breath 176 00:11:56,000 --> 00:11:59,079 Speaker 1: and hold the vowel sound ah like in the word father. 177 00:12:00,040 --> 00:12:02,360 Speaker 1: Please hold that sound for six seconds or until you 178 00:12:02,440 --> 00:12:10,760 Speaker 1: run out of breath, let's begin. Uh wof, I got 179 00:12:10,800 --> 00:12:14,520 Speaker 1: it done. That's a little bit harder than I thought. Okay, 180 00:12:14,559 --> 00:12:18,240 Speaker 1: my health score, I'm a low risk, I'm a fifty 181 00:12:19,000 --> 00:12:21,960 Speaker 1: means that today's score is in your normal range. And 182 00:12:22,000 --> 00:12:25,960 Speaker 1: now I'm getting my screening results. Sounds like you're feeling okay. 183 00:12:26,080 --> 00:12:28,680 Speaker 1: Watch for COVID nineteen symptoms and call your medical provider 184 00:12:28,720 --> 00:12:31,880 Speaker 1: if any symptoms develop. Follow your local and state public 185 00:12:31,880 --> 00:12:35,680 Speaker 1: health guidelines to keep yourself and others safe. So it 186 00:12:35,720 --> 00:12:38,200 Speaker 1: occurs to me, Laura, that part of the benefit of 187 00:12:38,200 --> 00:12:40,840 Speaker 1: a screening like this is that you just literally have 188 00:12:41,000 --> 00:12:44,679 Speaker 1: somebody every day walking through the process of thinking about 189 00:12:44,840 --> 00:12:49,280 Speaker 1: how they're feeling. Do I feel hot? Did I sleep poorly? 190 00:12:49,760 --> 00:12:53,360 Speaker 1: Am I losing my sense of taste or smell? Those 191 00:12:53,360 --> 00:12:55,520 Speaker 1: are things we should all be asking ourselves, but I 192 00:12:55,520 --> 00:12:58,720 Speaker 1: don't know that we do it intentionally every day. So 193 00:12:58,760 --> 00:13:00,959 Speaker 1: if you do have an app that's walking you through this, 194 00:13:01,559 --> 00:13:05,000 Speaker 1: that could be a big part of the benefit right there, 195 00:13:05,040 --> 00:13:08,280 Speaker 1: regardless of what you learn from your vocal bio markers. 196 00:13:11,640 --> 00:13:14,320 Speaker 1: That was Michelle fa Cortez and that's it for our 197 00:13:14,320 --> 00:13:17,520 Speaker 1: show today. For coverage of the outbreak from one twenty 198 00:13:17,559 --> 00:13:21,920 Speaker 1: bureaus around the world. Visit bloomberg dot com, slash Coronavirus, 199 00:13:22,840 --> 00:13:25,040 Speaker 1: and if you like the show, please leave us a 200 00:13:25,040 --> 00:13:28,960 Speaker 1: review and a rating on Apple Podcasts or Spotify. It's 201 00:13:29,000 --> 00:13:32,160 Speaker 1: the best way to help more listeners find our global reporting. 202 00:13:33,360 --> 00:13:37,360 Speaker 1: The Prognosis Daily edition is produced by tophor Forheas, Magnus 203 00:13:37,360 --> 00:13:41,960 Speaker 1: Hendrickson and me Laura Carlson. Today's main story was reported 204 00:13:41,960 --> 00:13:46,600 Speaker 1: by Michelle fay Cortez. Original music by Leo Sidran. Our 205 00:13:46,760 --> 00:13:50,800 Speaker 1: editors are Rick Shine and Francesco Leady. Francesco Levi is 206 00:13:50,840 --> 00:14:02,320 Speaker 1: Bloomberg's head of podcasts. Thanks for listening well.