Voice banking is the practice of recording a person’s natural voice so that, after they lose the ability to speak, a computer can generate new speech that still sounds like them. It matters most for people with conditions such as ALS (amyotrophic lateral sclerosis), where speech is progressively lost. Team Gleason, an ALS nonprofit, describes the core idea plainly: a person records a series of phrases, and software builds a “computerized” synthetic version of their natural voice from those recordings.
There are a few distinct techniques. Traditional voice banking has the person read a fixed script of phrases to build a synthetic voice. Message banking is different - it saves actual recordings of meaningful phrases (“I love you,” a laugh) in the person’s own voice for later playback, rather than synthesizing new sentences. The newest approach is AI voice cloning, which can build a synthetic voice from far less, sometimes from previously recorded audio like old voicemails or videos, and even “proxy” banking where a relative records phrases for someone who has already lost their voice. Team Gleason works with multiple providers and helps cover the cost.
The technology turns a generic robotic synthesizer into something that preserves identity. For someone facing the loss of their voice, continuing to “sound like themselves” in a communication device is not a cosmetic detail; it is a matter of who they are to their family. Clinicians stress doing it early, while the voice is still strong.
Why business readers should care: voice cloning is usually discussed as a fraud and deepfake risk, but the same capability has a clear humane use - preserving a person’s voice - and the contrast is a reminder that a technology’s ethics live in the use, not the model.