Inventor

SHARIFI MATTHEW

CH445 patents

⚠️ This page may combine multiple inventors who share the name “SHARIFI MATTHEW”. Patents are grouped by organization below to help distinguish them — per-person disambiguation is on the roadmap.

SHARIFI MATTHEW

GOOGLE INC

Hotword recognition

Providing an annotation linking related entities in onscreen content

Hotword detection on multiple devices

Hotword detection on multiple devices

Providing pre-computed hotword models

Initiating actions based on partial hotwords

Promoting voice actions to hotwords

Simulated hyperlinks on a mobile device based on user intent and a centered selection of text

Providing user assistance from interaction understanding

Indexing mobile onscreen content

Providing actions for mobile onscreen content

Replaying user input actions using screen capture images

Annotating relevant content in a screen capture image

Detection and ranking of entities from mobile onscreen content

Redaction suggestion for sharing screen content

Dual model speaker identification

Multi-channel audio video fingerprinting

Large-scale speaker identification

Speech endpointing based on voice profile

Identifying media content

Speaker identification using hash-based indexing

Efficient utterance-specific endpointer triggering for always-on hotwording

Synchronized content playback related to content recognition

Selectively obscuring private information based on contextual information

Search query predictions by a keyboard

Multiple speech locale-specific hotword classifiers for selection of a speech locale

Differentiating between near identical versions of a song

Initiating actions based on partial hotwords

Segment-based speaker verification using dynamically generated phrases

Drag-and-drop on a mobile device

IDF weighting of LSH bands for live reference ingestion

Segment-based speaker verification using dynamically generated phrases

Unified recognition of speech and music

GOOGLE LLC

Hotword detection on multiple devices

Speaker identification using a text-independent model and a text-dependent model

Systems and methods for automatically suggesting metadata for media content

Obtaining and enhancing metadata for content items

Hotword detection on multiple devices

Frequency based audio analysis using neural networks

Sharing screen content in a mobile environment

Screen capture image repository for a user

SHARIFI MATTHEW

Positioning using audio recognition

Full digest of an audio file for identifying duplicates

Detection of creative works on broadcast media

In-stream video stitching

Transformation invariant media matching

Learning common spelling errors through content matching

POSTELNICU GHEORGHE

Detection of inactive broadcasts during live stream ingestion

MATIAS YOSSI

Aggregation of related media content

WISEMAN MATTHEW

Inverted client-side fingerprinting and matching