Patents by Inventor Paolo Mairano

Paolo Mairano has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Systems and methods for generating speech of multiple styles from text
    Patent number: 10192541
    Abstract: A text-to-speech (TTS) system includes components capable of supporting the generation of speech output in any of multiple styles, and may switch seamlessly from producing speech output in one style to producing speech output in another style. For example, a concatenative TTS system may include a speech base storing speech units associated with multiple speech styles, and a linguistic analysis component to generate a phonetic transcription specifying speech output in any of multiple styles. Text input may include a style indication associated with a particular segment of the input text. The linguistic analysis component may invoke encoded rules and/or components based upon the style indication, and generate a phonetic transcription specifying a speech style, which may be processed to generate output speech.
    Type: Grant
    Filed: June 5, 2014
    Date of Patent: January 29, 2019
    Assignee: Nuance Communications, Inc.
    Inventors: Paolo Mairano, Corinne Bos-Plachez, Sourav Nandy, Johan Wouters, Silvia Maria Antonella Quazza, Dong-Jian Yue
  • SYSTEMS AND METHODS FOR GENERATING SPEECH OF MULTIPLE STYLES FROM TEXT
    Publication number: 20170186418
    Abstract: A text-to-speech (TTS) system includes components capable of supporting the generation of speech output in any of multiple styles, and may switch seamlessly from producing speech output in one style to producing speech output in another style. For example, a concatenative TTS system may include a speech base storing speech units associated with multiple speech styles, and a linguistic analysis component to generate a phonetic transcription specifying speech output in any of multiple styles. Text input may include a style indication associated with a particular segment of the input text. The linguistic analysis component may invoke encoded rules and/or components based upon the style indication, and generate a phonetic transcription specifying a speech style, which may be processed to generate output speech.
    Type: Application
    Filed: June 5, 2014
    Publication date: June 29, 2017
    Applicant: Nuance Communications, Inc.
    Inventors: Paolo Mairano, Corinne Bos-Plachez, Sourav Nandy, Johan Wouters, Silvia Maria Antonella Quazza, Dong-Jian Yue