METHOD AND SYSTEM FOR TEXT-TO-SPEECH SYNTHESIS WITH PERSONALIZED VOICE

• US 20150025891A1
• Filed: 10/10/2014
• Published: 01/22/2015
• Est. Priority Date: 03/20/2007
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Goldberg, Itzhack, Hoory, Ron, Mizrachi, Boaz, Kons, Zvi
• Granted Patent: US 9,368,102 B2
• Time in Patent Office: Days
• US Class Current: 704/260
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser
• CPC Class Codes: G10L 13/04 Details of speech synthesis systems, e.g. synthesiser structure or memory management

Abstract

A method and system are provided for text-to-speech synthesis with personalized voice. The method includes receiving an incidental audio input (403) of speech in the form of an audio communication from an input speaker (401) and generating a voice dataset (404) for the input speaker (401). The method includes receiving a text input (411) at the same device as the audio input (403) and synthesizing (312) the text from the text input (411) to synthesized speech including using the voice dataset (404) to personalize the synthesized speech to sound like the input speaker (401). In addition, the method includes analyzing (316) the text for expression and adding the expression (315) to the synthesized speech. The audio communication may be part of a video communication (453) and the audio input (403) may have an associated visual input (455) of an image of the input speaker. The synthesis from text may include providing a synthesized image personalized to look like the image of the input speaker with expressions added from the visual input (455).

Claims

1. A method for text-to-speech synthesis, comprising:

receiving, at a first device, incidental audio speech data over a first network communication link from a second device, wherein the incidental audio speech data is produced during an audio communication in which an operator of the second device participates and is representative of the operator'"'"'s speaking characteristics;

generating, by the first device, a voice dataset for the operator based, at least in part, on the incidental audio speech data;

receiving, at the first device, text data from the second device over a second network communication link subsequent to receiving the incidental audio speech data;

converting, by the first device, the text data to synthesized speech; and

personalizing the synthesized speech to sound like the operator using, at least in part, the voice dataset.