New York, August 1 (IANS): Teleconferencing platforms like Zoom and Microsoft Teams helped people stay connected during the coronavirus pandemic. But the platforms were unable to capture the sounds accurately enough for doctors to successfully treat and evaluate patients with voice and speech disorders, according to one study.
In a virtual world, voice therapy presents a unique challenge because physicians must rely on voice acoustic recordings to evaluate the effectiveness of their treatments. But many teleconferencing platforms distorted the sounds in their efforts to eliminate background noise, the study led by a team of researchers from Boston University in the US revealed.
As the pandemic developed and blockages moved a lot of online voice and speech therapy, “there was no consensus among clinicians (voice and speech) (who were) trying to become speech therapy. telepractice and we wanted to determine the accuracy of the acoustic measurements that “telepractice” can do, said Hasini Weerathunge, a graduate student at BU’s Rafik B Hariri Institute for Computational and Computer Science and Engineering.
The team tested five different HIPAA-compliant teleconferencing platforms: Cisco Webex, Microsoft Teams, Doxy.me, VSee Messenger, and Zoom.
In a soundproofed room, the team recorded voice samples from 29 patients, ages 18 to 82, who had a variety of voice or voice diagnoses. These recordings were reproduced to researchers via an external speaker via teleconferencing platforms, simulating telepractice conversations.
The team found that each platform has its own audio enhancement algorithm that affects sound quality. Zoom was the only platform that allowed users to disable these audio enhancement features, which allowed researchers to test the platform’s original audio.
All teleconferencing platforms did a poor job of capturing many of the measures needed for accurate and clinically meaningful voice assessments. The pitch varied significantly across all virtual platforms compared to real-life recordings. The researchers said this could be due to internet connection or bandwidth issues affecting the way and time sounds are transmitted across platforms.
They also found that the dynamic range of vocal sonority measured during telepractice was very different from live recordings.
The effect was even true for Zoom, where researchers could turn off audio enhancements.
Overall, “Microsoft Teams had the best performance, as all of our voice measurements were the least affected on this platform,” Weerathunge said.
Because many of the voice metrics collected from virtual platforms showed clinically significant differences from in-person collections, Weerathunge and the team call caution for voice and speech therapists using telepractice.