0.5. Generation

The story of Stephen Hawking is perhaps one of the most remarkable stories of all time. A brilliant mathematician, he was diagnosed with Amyolateral Sclerosis (ALS) at just age 21. Having just started graduate school, he was given 2 years to live.

Despite this, he eventually finished his thesis in the area of general relativity and cosmology - and went on to be one of the most prolific physicists of our time.

None of this would have been possible without speech-generating voice computing devices. At first, he could generate only a few words. Then he could generate 15 words per minute through a primitive computer program with an American accent. As his condition worsened and older versions became unusable (due to more restrictions in muscle movement), newer versions of speech synthesizers were created - using brain waves (EEGs) and facial expressions to communicate speech. These primitive tools allowed for Stephen Hawking to continue to contribute to the scientific community up until his death (2018).

This story shows you the power of voice computing. Not only can they help restore functionality to those who have lost their voices, it opens up a whole new way for us to augment our own voices as they are now - create clones of our own voices, generate new vocabularies and transcription models, and even generate new poems. In other words, voice computers can emit a sense of individuality and creativity in how we interact with others in this ever-changing world.

This chapter is all about using computing platforms to automate the creation of voice content.

Specifically, it will cover:

• 5.1 - Machine-generated voice data
• 5.2 - Generating text data
• 5.3 - Generating audio data
• 5.4 - Generating mixed data

In this way, you can harness the power of large datasets for generative purposes!