AphaVoice generates realistic aphasia speech for medical training
AphaVoice: Text-to-Speech Model for Aphasia Patient Simulation
Computation and LanguageArtificial Intelligence
Summary
People recovering from strokes sometimes have trouble speaking clearly, which makes it hard for doctors to practice helping them. The authors created a computer system called AphaVoice that can read text aloud in a way that mimics the speech patterns of people with aphasia. This system sounds realistic by capturing speech disruptions and keeping the speaker's voice consistent. It can even produce different speaker voices while still sounding like someone with aphasia. This could help train medical professionals more effectively.
What this means in practice
- •For speech therapists: Use realistic aphasic speech to train in diagnosis and therapy techniques without needing actual patient recordings.
- •For medical simulation developers: Create scalable training tools featuring diverse aphasic speech patterns that improve clinical education quality.$Commercial implications: Enables developing and selling advanced clinical simulators for aphasia that use realistic synthetic speech.
Authors
Al Varrel Putra Kusuma, Fernando Bello, Joshua Brown
Abstract
While language therapy is the primary treatment for post-stroke aphasia, limited clinical resources often restrict patient access to proper care. Addressing this gap requires reliable training simulations for medical professionals. However, current AI efforts focus almost exclusively on speech processing, leaving speech generation largely unexplored. This study introduces AphaVoice, a novel text-to-speech (TTS) framework explicitly engineered to synthesize highly realistic aphasic speech. Evaluated against state-of-the-art TTS systems, AphaVoice demonstrates a superior ability to capture the clinical feature of dysfluency, outperforming existing models in temporal metrics while consistently maintaining speaker identity. By leveraging latent space shifting, the model also supports multi-speaker generation without sacrificing phonetic realism. Ultimately, AphaVoice provides a scalable, high-fidelity tool for medical training and clinical simulation related to aphasia.