Meryem P Donmez

20130158981, Jun 20, 2013

Dec 20, 2011

13/331215

Hakan Ceylan - San Francisco CA, US
Ergin Elmacioglu - San Jose CA, US
Meryem Pinar Donmez - San Francisco CA, US

Yahoo! Inc. - Sunnyvale CA

G06F 17/27

704 9, 704E11001

Methods, systems, and computer programs are presented for linking newsworthy events in a document to published content. One method includes an operation for receiving features by a classifier that is operable to determine a probability of the availability of news for a sentence. When the features are found in the sentence, the probability of the availability of news for the sentence increases, where the sentence includes one or more noun phrases and ends in a full stop. The classifier determines which sentences in a document are candidate sentences for being linked to news articles, and for each candidate sentence, the method includes an operation for finding an associated news article when there is an associated news article exceeding a relevance threshold. Further, the method includes operations for adding links in the document to the found associated news articles, and for displaying the document with the added links.

20220188361, Jun 16, 2022

Dec 11, 2020

17/120013

- Menlo Park CA, US
Nanshu Wang - Sunnyvale CA, US
Fan Wang - Sunnyvale CA, US
Meryem Pinar Donmez Ediz - Mountain View CA, US
Omer Muzaffar - Bothell WA, US
Kshitiz Malik - Palo Alto CA, US
Vikas Seshagiri Rao Bhardwaj - Kirkland WA, US
Anuj Kumar - Santa Clara CA, US
Shreyan Bakshi - Pittsburgh PA, US

G06F 16/9032
G10L 15/16
G10L 15/06
H04L 12/58
G02B 27/01

In one embodiment, a method includes receiving a first input by a user from a client system associated with the user, wherein the first input is in a voice modality, analyzing the first input to generate one or more candidate hypotheses, determining one or more modalities for presenting output generated by the one or more computing systems to the user at the client system, and sending instructions to the client system for presenting one or more suggested auto-completions corresponding to one or more of the candidate hypotheses, respectively, wherein each suggested auto-completion comprises the corresponding candidate hypothesis, and wherein the one or more suggested auto-completions are presented in the one or more determined modalities.

20180330725, Nov 15, 2018

Aug 18, 2017

15/681197

- Redmond WA, US
Shuangyu CHANG - Fremont CA, US
Khuram SHAHID - Seattle WA, US
Meryem Pinar DONMEZ EDIZ - Los Altos CA, US
Nitin AGARWAL - Fremont CA, US

G10L 15/19
G10L 15/18
G10L 15/26

A method for priming an extensible speech recognition system comprises receiving audio language input from a user. The method also comprises receiving an indication that the audio language input is associated with a first language-based intelligent agent. The first language-based intelligent agent is associated with a first grammar set that is specific to the first language-based intelligent agent. Additionally, the method comprises matching one or more spoken words or phrases within the audio language input to text-based words or phrases within a general grammar set associated with a speech recognition system and the first grammar set. The first grammar set is associated with a higher match bias than the general grammar set, such that the speech recognition system is more likely to match the one or more spoken words or phrases to the text-based words or phrases within the first grammar set.