Google Inc.

Cisco Technology, Inc.

Microsoft Technology Licensing, LLC

Nuance Communications, Inc.

Moxair, Inc.

Sri International

Ananth Sankar

Navdeep Jaitly

Christopher S Co

Katherine Irene Chou

Lily Hao Yi Peng

Francoise Beaufays

Anurag Bist

Sachin S Kajarekar

Mitchel Weintraub

Shaun Eric Williams

David Romacho Rosell

Jim Chen Chou

Matthias Paulik

Satish K Gannu

Praveen Dua

Vipin Gupta

Saurabh Kataria

Ying Xuan

Venkata Ramana Rao Gadde

Sumit Srivastava

Suhit Sinha

Sriram Sundararajan

Bharat Kumar Jain

Vivek Halder

Bhargavkumar Kanubhai Patel

Sirjan Kafle

Di Wen

Sakshi Verma

Xue Xia

Nikita Gupta

Xun Luan

Aman Gupta

Methods, systems, and apparatus, including computer programs encoded on computer storage media for speech recognition. One method includes obtaining an input acoustic sequence, the input acoustic sequence representing one or more utterances; processing the input acoustic sequence using a speech recognition model to generate a transcription of the input acoustic sequence, wherein the speech recognition model comprises a domain-specific language model; and providing the generated transcription of the input acoustic sequence as input to a domain-specific predictive model to generate structured text content that is derived from the transcription of the input acoustic sequence.

Generating structured text content using speech recognition models

Described herein are systems and methods for generating an embedding—a learned representation—for an image. The embedding for the image is derived to capture visual aspects, as well as textual aspects, of the image. An encoder-decoder is trained to generate the visual representation of the image. An optical character recognition (OCR) algorithm is used to identify text/words in the image. From these words, an embedding is derived by performing an average pooling operation on pre-trained embeddings that map to the identified words. Finally, the embedding representing the visual aspects of the image is combined with the embedding representing the textual aspects of the image to generate a final embedding for the image.

Utilizing visual and textual aspects of images with recommendation systems

Methods, systems, and computer programs are presented for detecting near duplicates and partial matches of videos. One method includes an operation for receiving a video containing frames. For each frame, keypoints are determined within the frame. For each keypoint, a horizontal gradient vector is calculated based on a horizontal gradient at the keypoint and a vertical gradient vector is calculated based on a vertical gradient at the keypoint. The horizontal and vertical gradients are binary vectors. Further, a keypoint description is generated for each keypoint based on the horizontal gradient vector and the vertical gradient vector. Further, the frames are matched to frames of videos in a video library based on the keypoint descriptions of the keypoints in the frame in the videos in the video library. Further, a determination is made if the video has near duplicates in the video library based on the matching.

Partial-video near-duplicate detection

Speech audio that is intended for transcription into textual form is received. The received speech audio is divided into first speech segments. A plurality of speakers is identified. A speaker is configured for repeating in spoken form a first speech segment that the speaker has listened to. A subset of speakers is determined for sending each first speech segment. Each first speech segment is sent to the subset of speakers determined for the particular first speech segment. The second speech segments are received from the speakers. The second speech segment is a re-spoken version of a first speech segment that has been generated by a speaker by repeating in spoken form the first speech segment. The second speech segments are processed to generate partial transcripts. The partial transcripts are combined to generate a complete transcript that is a textual representation corresponding to the received speech audio.

Crowd sourcing audio transcription via re-speaking

A method is provided and includes estimating an approximate list of potential speakers in a file from one or more applications. The file (e.g., an audio file, video file, or any suitable combination thereof) includes a recording of a plurality of speakers. The method also includes segmenting the file according to the approximate list of potential speakers such that each segment corresponds to at least one speaker; and recognizing particular speakers in the file based on the approximate list of potential speakers.

Speaker segmentation and recognition based on list of speakers

An example method is provided and includes receiving a video bitstream in a network environment; detecting a question in a decoded audio portion of a video bitstream; and marking a segment of the video bitstream with a tag. The tag may correspond to a location of the question in the video bitstream, and can facilitate consumption of the video bitstream. The method can further include detecting keywords in the question, and combining the keywords to determine a content of the question. In specific embodiments, the method can also include receiving the question and a corresponding answer from a user interaction, crowdsourcing the question by a plurality of users, counting a number of questions in the video bitstream and other features.

System and method for question detection based video segmentation, search and collaboration in a video processing environment

In visual display devices such as LCD devices with backlight illumination, the backlight typically consumes most of device battery power. In the interest of displaying a given pixel pattern at a minimized backlight level, the pattern can be transformed while maintaining image quality, with a transform determined from pixel luminance statistics. Aside from, or in addition to being used for such minimizing, a transform also can be used for image enhancement, for a displayed image better to meet a visual perception quality. In either case, the transform preferably is constrained for enforcing one or several display attributes. In a network setting, the technique can be implemented in distributed fashion, so that subtasks of the technique are performed by different, interconnected processors such as server, client and proxy processors.

Distributed processing for video enhancement and display power management

In visual display devices such as LCD devices with backlight illumination, the backlight typically consumes most of device battery power. In the interest of displaying a given pixel pattern at a minimized backlight level, the pattern can be transformed while maintaining image quality, with a transform determined from pixel luminance statistics. Aside from, or in addition to such minimizing, a transform also can be used for image enhancement, for a displayed image better to meet a visual perception quality. In either case, the transform preferably is constrained for enforcing one or several display attributes.

Video enhancement and display power management

A computerized method is provided for generating pronunciations for words and storing the pronunciations in a pronunciation dictionary. The method includes graphing sets of initial pronunciations; thereafter in an ASR subsystem determining a highest-scoring set of initial pronunciations; generating sets of alternate pronunciations, wherein each set of alternate pronunciations includes the highest-scoring set of initial pronunciations with a lowest-probability phone of the highest-scoring initial pronunciation substituted with a unique-substitute phone; graphing the sets of alternate pronunciations; determining in the ASR subsystem a highest-scoring set of alternate pronunciations; and adding to a pronunciation dictionary the highest-scoring set of alternate pronunciations.

Method for learning linguistically valid word pronunciations from acoustic data

A computerized pronunciation system is provided for generating pronunciations for words and storing the pronunciations in a pronunciation dictionary. The system includes a word list including at least one word; transcribed acoustic data including at least one waveform for the word and transcribed text associated with the waveform; a pronunciation-learning module configured to accept as input the word list and the transcribed acoustic data, the pronunciation-learning module including: sets of initial pronunciations of the word, a scoring module configured score pronunciations and to generate phone probabilities, and a set of alternate pronunciations of the word, wherein the set of alternate pronunciations include a highest-scoring set of initial pronunciations with a highest-scoring substitute phone substituted for a lowest-probability phone; and a pronunciation dictionary configured to receive the highest-scoring set of initial pronunciations and the set of alternate pronunciations.

Method and system for learning linguistically valid word pronunciations from acoustic data

Probabilistic recognition using clusters and simple probability functions provides improved performance by employing a limited number of clusters each using a relatively large number of simple probability functions. The simple probability functions for each of the limited number of state clusters are greater in number than the limited number of state clusters.

Method and apparatus for probabilistic recognition using small number of state clusters

Growing community of inventors

Palo Alto, CA, United States of America

Ananth Sankar