All articles tagged with #voice analysis
The podcast discusses how AI is being used to analyze voices for detecting diseases like cancer, Parkinson's, and Alzheimer's, highlighting the potential of voice as a non-invasive biomarker for health assessment and diagnosis, with future applications in telehealth and everyday health monitoring.
Researchers have developed machine learning algorithms that can detect subtle vocal changes associated with vocal fold lesions, potentially enabling earlier diagnosis of laryngeal cancer through non-invasive voice recordings, especially in men, with future applications possibly extending to women and broader clinical use.
Researchers at Arizona State University found that OpenAI's discontinued GPT-4o voice, "Sky," sounds very similar to Scarlett Johansson's voice, though not identical. The analysis showed Johansson's voice was more similar to Sky than 98% of other actresses, but Anne Hathaway and Keri Russell's voices sometimes resembled Sky's more. OpenAI's CEO and CTO have denied any intention to mimic Johansson's voice, though Johansson has hired legal counsel and may take legal action.
Researchers at Arizona State University found that OpenAI's now-pulled "Sky" voice assistant bears a striking resemblance to Scarlett Johansson's voice, more so than 98% of other actresses analyzed. Despite OpenAI's claims that the similarity was unintentional, Johansson is outraged, especially since OpenAI's CEO Sam Altman had previously reached out to her to license her voice. The debate continues over whether the resemblance was a deliberate design choice or a coincidence.
Researchers from Klick Applied Sciences in Toronto have developed an artificial intelligence model that uses a 10-second voice recording to predict diabetes risk. The AI program was shown to predict the disease with 85% accuracy, using voice features that are affected by complications of type 2 diabetes such as peripheral neuropathy and muscle weakness. The voice changes associated with type 2 diabetes differed between men and women. The researchers aim to make this technology widely available to increase accessibility and early detection of type 2 diabetes. However, the accuracy and reliability of AI in detecting diabetes from voice alone need rigorous validation against current diagnostic methods, and data privacy concerns must be addressed.
Researchers from Klick Labs have developed an artificial intelligence (AI) model that can detect type 2 diabetes by analyzing a person's voice. The AI tool, combined with basic health information, was 89% accurate in diagnosing type 2 diabetes in women and 86% accurate in men. Changes in vocal characteristics, such as pitch and strength, can indicate diabetic neuropathy or nerve damage associated with diabetes. While more research is needed to validate the effectiveness of this method across different populations and factors, the AI voice test shows promise as a non-intrusive and accessible screening tool for diabetes.