Hu U Peng

US Patent:

7386451, Jun 10, 2008

Filed:

Sep 11, 2003

Appl. No.:

10/660388

Inventors:

Min Chu - Beijing, CN
Hu Peng - San Francisco CA, US
Yong Zhao - Beijing, CN

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G10L 13/08
G10L 13/00

US Classification:

704260, 704258

Abstract:

A method is provided for optimizing an objective measure used to estimate mean opinion score or naturalness of synthesized speech from a speech synthesizer. The method includes using an objective measure that has components derived directly from textual information used to form synthesized utterances. The objective measure has a high correlation with mean opinion score such that a relationship can be formed between the objective measure and corresponding mean opinion score. The objective measure is altered to provide a different function of textual information derived from the utterances so as to improve the relationship between the scores of the objective measure and subjective ratings of the synthesized utterances.

US Patent:

20040148171, Jul 29, 2004

Filed:

Sep 15, 2003

Appl. No.:

10/662985

Inventors:

Min Chu - Beijing, CN
Hu Peng - San Francisco CA, US
Yong Zhao - Beijing, CN

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G10L013/00

US Classification:

704/258000

Abstract:

A speech synthesizer is provided that concatenates stored samples of speech units without modifying the prosody of the samples. The present invention is able to achieve a high level of naturalness in synthesized speech with a carefully designed training speech corpus by storing samples based on the prosodic and phonetic context in which they occur. In particular, some embodiments of the present invention limit the training text to those sentences that will produce the most frequent sets of prosodic contexts for each speech unit. Further embodiments of the present invention also provide a multi-tier selection mechanism for selecting a set of samples that will produce the most natural sounding speech.

US Patent:

20050119891, Jun 2, 2005

Filed:

Jan 6, 2005

Appl. No.:

11/030208

Inventors:

Min Chu - Beijing, CN
Hu Peng - San Francisco CA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G10L013/08

US Classification:

704260000

Abstract:

A speech synthesizer is provided that concatenates stored samples of speech units without modifying the prosody of the samples. The present invention is able to achieve a high level of naturalness in synthesized speech with a carefully designed training speech corpus by storing samples based on the prosodic and phonetic context in which they occur. In particular, some embodiments of the present invention limit the training text to those sentences that will produce the most frequent sets of prosodic contexts for each speech unit. Further embodiments of the present invention also provide a multi-tier selection mechanism for selecting a set of samples that will produce the most natural sounding speech.

US Patent:

20210084114, Mar 18, 2021

Filed:

Nov 23, 2020

Appl. No.:

17/102075

Inventors:

- San Mateo CA, US
Hu Peng - Orinda CA, US
Stephanie Fok - San Francisco CA, US

International Classification:

H04L 29/08
H04L 12/26

Abstract:

The present disclosure describes methods and system for personalizing user experience in a gaming network. In particular, the user experience is personalized based on how users are characterized based on their interactions with the gaming network. By using the characterizations, the users are assigned labels that publicly identify the user experience, interest, and proficiency with various video games. Communities are then established based on users having the same labels. Lastly, user generated content can be characterized and promoted based on the characterizations to the appropriate communities.

US Patent:

20190351335, Nov 21, 2019

Filed:

May 18, 2018

Appl. No.:

15/984058

Inventors:

- San Mateo CA, US
Hu Peng - Orinda CA, US

International Classification:

A63F 13/86
A63F 13/53
A63F 13/52

Abstract:

The present disclosure describes methods and systems directed towards implementing scene tagging. In particular, scene tagging is used to characterize user generated content associated with gameplay information associated with video games. Exemplary user generated content include video recordings or screenshots from within the video game that pertain to events that have occurred. Corresponding metadata would then be used to describe substantive aspects about the user-generated content such as details about where the recording or screenshot was taken, who was included in the user-generated content, and what objects are found in the user-generated content. The associated metadata would 1) allow users to search among various uploaded user generated content and 2) search within portions of those use-generated content for particular sub-sections of interest. Furthermore, the metadata associated with the user-generated content would allow viewers to experience the same events that were recorded.

US Patent:

20190356753, Nov 21, 2019

Filed:

May 18, 2018

Appl. No.:

15/983948

Inventors:

- San Mateo CA, US
Hu Peng - Orinda CA, US
Stephanie Fok - San Francisco CA, US

International Classification:

H04L 29/08
H04L 12/26

Abstract:

The present disclosure describes methods and system for personalizing user experience in a gaming network. In particular, the user experience is personalized based on how users are characterized based on their interactions with the gaming network. By using the characterizations, the users are assigned labels that publicly identify the user experience, interest, and proficiency with various video games. Communities are then established based on users having the same labels. Lastly, user generated content can be characterized and promoted based on the characterizations to the appropriate communities.