Montreal Woman With ALS Hears Her Own Voice Again Thanks to AI-Powered Speech App

Teslim Teslim
2 Min Read

MONTREAL — When Anne-Marie Poirier speaks at family gatherings, her nieces and nephews hear a familiar voice—one they thought they might never hear again.

Poirier, who was diagnosed with ALS in 2023, has lost much of her ability to speak as the neurodegenerative disease progresses. But a new AI-powered app has given her back a version of her own voice, synthesized from recordings made before her diagnosis.

The app, developed by a Montreal-based tech company in partnership with researchers at a local university, uses machine learning to recreate a person’s vocal identity from a relatively small sample of speech. For Poirier, that meant recording a series of sentences and phrases while she still had the ability to speak clearly.

“When I heard it for the first time, I cried,” Poirier said through the app, which speaks on her behalf. “It sounded like me. Not a robot. Me.”

Her sister, Marie-Claude Poirier, said the change has been profound for the family. “The kids were scared at first—they didn’t understand why Aunt Anne-Marie sounded different. Now they run to her when she talks. It’s like she’s back.”

The technology is part of a growing field of AI-assisted communication tools for people with ALS and other conditions that affect speech. While text-to-speech has existed for decades, the ability to recreate a person’s unique vocal qualities—tone, pitch, rhythm, accent—is relatively new.

The company behind the app says it has worked with more than 200 patients in Quebec and Ontario since launching the pilot program last year. The service is provided free to patients, with funding from provincial health authorities and private donors.

For Poirier, the app has restored more than just her voice. “I can still be myself,” she said. “I can still tell my nieces I love them. That matters.”


Share This Article
Leave a Comment