Microsoft partners with Project Gutenberg to use AI to create 5,000 free audiobooks

IT House News on September 12, Microsoft recently cooperated with the "Free Digital Library Project" Project Gutenberg to use AI to produce more than 5,000 free audiobooks for the e-book library.

▲ Picture source Project Gutenberg website

According to reports, in this project, researchers used a variety of artificial intelligence technologies such as machine learning, automatic text selection, and natural speech synthesis to successfully realize the entire process of automatically converting e-books into speech

IT Home learned from the inquiry that the project team first developed an algorithm that can automatically analyze the structure of e-books, distinguish the main text and non-key parts of page numbers, tables, etc., and then use WaveNet, Tacotron and FastSpeech to convert text Speech technology generates natural reading speech similar to that of a real person. In addition, the team also enabled the system to identify the narrator, dialogue and different characters in the book, and adjust the voice accordingly to achieve the voice expression of the storyline.

Microsoft said that they have collected and sorted out more than 35,000 hours of audiobook voice data, including various classic literature, dramas, biographies and other works. This voice data will be freely available as an open source. Users can also use a small amount of their own recorded voice to generate the effect of "reading the entire book in their own voice." This project enriches the audiobook content of the website and also provides convenience for the visually impaired

To rewrite the content without changing the original meaning, the language needs to be changed to Chinese. The original sentence

does not need to appear

The above is the detailed content of Microsoft partners with Project Gutenberg to use AI to create 5,000 free audiobooks. For more information, please follow other related articles on the PHP Chinese website!