Google Inc.

Allgo Embedded Systems Private Limited

Abhay Sharma

Manish Kumar

Abhirut Gupta

Satendra Pal Singh

Kiran Rajaram

Aravindan Raghuveer

Sachin Sudhakar Farfade

Nitin Raut

Systems and methods for generating phonetic spelling variations of a given word based on locale-specific pronunciations. A phoneme-letter density model may be configured to identify a phoneme sequence corresponding to an input word, and to identify all character sequences that may correspond to an input phoneme sequence and their respective probabilities. The phoneme-phoneme error model may be configured to identify locale-specific alternative phoneme sequences that may correspond to a given phoneme sequence, and their respective probabilities. Using these two models, a processing system may be configured to generate, for a given input word, a list of alternative character sequences that may correspond to the input word based on locale-specific pronunciations, and/or a probability distribution representing how likely each alternative character sequence is to correspond to the input word.

Systems and methods for generating locale-specific phonetic spelling variations

The proposed Dynamic Motion Vector Analysis method applies to the motion compensation module of a video decoder system. The method analyzes the motion-vectors of a given region of picture frame and outputs a set of regions to be fetched from the reference frames stored in the external memory. The size and number of regions are decided by a hierarchical method that uses a set of user-defined input thresholds. Pre-processing of the motion vectors associated with the given region allows the method to handle reference data to be fetched from multiple reference frames in the same framework. A complementary dynamic batch (region of operation) size strategy that works along with MV-analysis is also proposed to help utilize the on-chip memory resources more efficiently.

Dynamic motion vector analysis method

Growing community of inventors

Bengaluru, India

Abhay Sharma