G10L 17/18
Artificial neural networks
Official title: Artificial neural networks; Connectionist approaches
G10L17/18 (Artificial neural networks) appears on 474 US grants and 669 published applications in this corpus.
Search patents in G10L- Granted patents
- 474
- Published applications
- 669
- Assignees
- 248
- Inventors
- 1,360
- Latest grant
- 6 Oct 2026
Filings by year
| Year | Granted patents | Published applications |
|---|---|---|
| 1995 | 1 | 0 |
| 1996 | 1 | 0 |
| 1997 | 0 | 0 |
| 1998 | 1 | 0 |
| 1999 | 0 | 0 |
| 2000 | 0 | 0 |
| 2001 | 1 | 0 |
| 2002 | 0 | 0 |
| 2003 | 1 | 0 |
| 2004 | 0 | 0 |
| 2005 | 0 | 0 |
| 2006 | 0 | 0 |
| 2007 | 0 | 0 |
| 2008 | 0 | 0 |
| 2009 | 0 | 0 |
| 2010 | 0 | 1 |
| 2011 | 0 | 2 |
| 2012 | 0 | 0 |
| 2013 | 1 | 0 |
| 2014 | 0 | 2 |
| 2015 | 2 | 8 |
| 2016 | 8 | 10 |
| 2017 | 5 | 13 |
| 2018 | 13 | 34 |
| 2019 | 21 | 33 |
| 2020 | 37 | 60 |
| 2021 | 43 | 83 |
| 2022 | 59 | 95 |
| 2023 | 61 | 89 |
| 2024 | 84 | 100 |
| 2025 | 82 | 66 |
| 2026 | 53 | 73 |
By publication year. Applications have been published since 2001. 2026 runs to 8 Oct.
Recent patents
- Authentication using active acoustic sensing6 Oct 2026US 12,754,987 B2Google Inc.
- Using machine learning and discrete tokens to estimate different sound sources from audio mixtures22 Sep 2026US 12,744,045 B2Google Inc.
- Audio authentication for digital telephony1 Sep 2026US 12,726,477 B1Zoom Communications, Inc.
- Learning data generating device, learning data generating method, learning device, learning method, data structure, information processor, and acoustic treatment device25 Aug 2026US 12,718,149 B2Sony Group Corporation
- Method and apparatus for speaker verification based on next-TDNN25 Aug 2026US 12,718,821 B2Hyundai Motor Company
- Training and deployment of machine learning models8 Oct 2026US 2026/0310962 A1NVIDIA Corporation
- Cross-lingual prosodic voice cloning in plurality of languages8 Oct 2026US 2026/0311715 A1Camb Ai, Inc.
- Line-of-credit processing method and apparatus, and computer device and computer program product1 Oct 2026US 2026/0301064 A1Shenzhen Tcl New Technology Co., Ltd.
- Joint Acoustic Echo Cancellation (AEC) and Personalized Noise Suppression (PNS)1 Oct 2026US 2026/0301758 A1Microsoft Technology Licensing, LLC
- Multi-layer machine-learning adversarial threat defense in voice and chat channels1 Oct 2026US 2026/0303722 A1
Most cited
By later US patents
- End-to-end speaker recognition using deep neural network130US 9,824,692 B1Pindrop Security, Inc. · 2017
- Model adaptation of neural tree networks and other fused models for speaker verification99US 6,519,561 B1T-Netix, Inc. · 2003
- Speaker verification system using decision fusion logic97US 5,839,103 ARutgers, The State University of New Jersey · 1998
- Multi-microphone speech separation82US 10,957,337 B2Microsoft Technology Licensing, LLC · 2021
- Permutation invariant training for talker-independent multi-talker speech separation71US 10,249,305 B2Microsoft Technology Licensing, LLC · 2019