Peer-reviewed research
Reidentification of Participants in Shared Clinical Data Sets: Experimental Study
Daniela Wiepert et al., JMIR AI, 2024
Read the originalRead on October 17, 2026
- What it is
- An experimental study that tests whether speaker identification models can match people to their recordings in shared clinical speech data sets, without demographic or metadata cues. Its background section explains why voice itself can act as an identifier.
- How much weight it carries
- Peer-reviewed and open access. The experiments use clinical recordings from people with speech disorders, so the measured re-identification risk applies to that setting and may not transfer to everyday dictation.
Facts taken from this source
- The authors of a 2024 study say that speech can potentially identify speakers, so sharing voice recordings raises privacy concerns.
“As speech can potentially identify speakers (ie, voiceprints), sharing recordings raises privacy concerns.” (Abstract (Background))