Voice Clone Platform Debuts ‘Voice Bank’ for Future Speech Synthesis
AI voice cloning platform VoCapsule has introduced a service for recording and securely storing voice recordings that could be used to generate a synthetic version of their voice in the future. The new My Legacy Voice platform manages the ‘voice banking’ and provides access to the recordings and a voice clone to whoever the user wishes and counts legendary rapper Ice-T as a new client and brand ambassador. The company pitches the service as a way to leave something to loved ones after passing away or for those who may lose the ability to speak due to ALS or other degenerative diseases.
My Legacy Voice is set up as a relatively streamlined path for becoming a member and making and storing the audio recordings necessary to train an AI model to imitate how they sound. Once the recordings are complete, VoCapsule will store the data in its archive, sealing it away from anyone except a designated primary recipient of the data should the member pass away or develop a neurological condition. The original member can also access the data if they start to lose the ability to speak and want to make a voice clone. VoCapsule’s natural language processing and AI models are designed to realistically imitate the way someone speaks, including emotional tone, accent, and cadence. The company claims as much as 99% ‘psychoacoustic accuracy,’ referring to how close to a human the synthetic voice comes.
“We understand a major problem for those who have lost someone they love,” VoCapsule CEO Brock Daily said. “Nobody should ever say, ‘I wish I could hear their voice again.'”
Bringing back the voices of loved ones who have passed away or who may no longer be able to speak is a growing aspect of the synthetic media market. South Korean virtual human developer DeepBrain AI recently began producing AI memorial services for funerals with synthetic recreations of the voice and face of someone who has passed away, something synthetic video startup Storyfile has also begun testing. And Amazon showcased an upcoming synthetic speech feature for Alexa with a video of a child having the voice assistant read a story to him in the voice of his deceased grandmother. VoCapsule’s ability to replicate the way specific humans speak will give it a good shot to carve a significant portion of the market judging from Ice-T’s enthusiasm about it.
“I played my voice clone and my son didn’t even recognize it as not being real, and from that point on I’ve been all in.”Ice-T said of his synthetic counterpart. “VoCapsule is going to be a game changer. I really see this as the future.”