Empower Conference: Automatic Speech Recognition & Voice Cloning

Change or loss of natural voice is likely for many people living with progressive neurological conditions such as MND/ALS, Parkinson’s Disease, MSA, PSP and many others. Also, for people living with the effects of head and neck cancer, trauma. And for many others, they may have never had natural voice to communicate with, for example some people with cerebral palsy. Is this situation, text to speech (or picture to speech) apps are sometimes used, using a synthesised voice that frequently sounds very different to how a person sounded or should be represented. 

The alternative is to create a ‘voice bank’ – a synthetic version of a person’s own voice from recordings or natural speech. However, this technology is limited to a few European languages, with none for languages spoken in Asia, and frequently very expensive. The quality is often robotic and easily distinguishable from a person’s natural voice. 

ElevenLabs is a new type of voice banking (often called voice cloning). It is highly accurate in reflecting how a person’s voice sounds. It is available in 73 languages including Assamese, Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi, Nepali, Punjabi, Sindhi, Tamil, Telugu. More are coming. It is free for people living with the conditions mentioned above. 

The workshop will demonstrate ElevenLabs technology end to end, creating voice banks in real time with plenty of time to discuss your questions and to try it out. The goal is to raise awareness and meet with organizations that support the conditions listed above – and others where voice change and loss is likely. ElevenLabs is a free option for people in India at risk of voice loss to still sound like themselves.