G10L 15/146
With insufficient amount of training data
Official title: {with insufficient amount of training data, e.g. state sharing, tying, deleted interpolation}
G10L15/146 (With insufficient amount of training data) appears on 7 US grants and 6 published applications in this corpus.
Search patents in G10L- Granted patents
- 7
- Published applications
- 6
- Assignees
- 8
- Inventors
- 20
- Latest grant
- 11 Jun 2024
Filings by year
| Year | Granted patents | Published applications |
|---|---|---|
| 1997 | 1 | 0 |
| 1998 | 2 | 0 |
| 1999 | 0 | 0 |
| 2000 | 0 | 0 |
| 2001 | 0 | 0 |
| 2002 | 1 | 0 |
| 2003 | 0 | 0 |
| 2004 | 0 | 0 |
| 2005 | 0 | 1 |
| 2006 | 0 | 0 |
| 2007 | 0 | 2 |
| 2008 | 0 | 0 |
| 2009 | 0 | 0 |
| 2010 | 0 | 0 |
| 2011 | 0 | 0 |
| 2012 | 0 | 0 |
| 2013 | 0 | 0 |
| 2014 | 0 | 0 |
| 2015 | 0 | 0 |
| 2016 | 0 | 1 |
| 2017 | 1 | 0 |
| 2018 | 0 | 0 |
| 2019 | 0 | 0 |
| 2020 | 0 | 0 |
| 2021 | 0 | 1 |
| 2022 | 1 | 1 |
| 2023 | 0 | 0 |
| 2024 | 1 | 0 |
By publication year. Applications have been published since 2001.
Recent patents
- Generating event output11 Jun 2024US 12,008,995 B2Amazon Technologies, Inc.
- Generating event output3 May 2022US 11,322,150 B2Amazon Technologies, Inc.
- Out of vocabulary pattern learning28 Mar 2017US 9,607,618 B2Nice Systems Ltd.
- Discriminatively trained mixture models in continuous speech recognition3 Dec 2002US 6,490,555 B1ScanSoft, Inc.
- Method and apparatus for speech recognition using optimized partial mixture tying of HMM state functions20 Oct 1998US 5,825,978 ASRI International
- Generating event output10 Nov 2022US 2022/0358930 A1Amazon Technologies, Inc.
- Generating event output29 Jul 2021US 2021/0233534 A1Amazon Technologies, Inc.
- Out of vocabulary pattern learning16 Jun 2016US 2016/0171973 A1Nice Systems Ltd.
- System and method for generating heterogeneously tied gaussian mixture models for automatic speech recognition acoustic models8 Nov 2007US 2007/0260459 A1Texas Instruments Incorporated
- System and method for creating generalized tied-mixture hidden Markov models for automatic speech recognition8 Feb 2007US 2007/0033044 A1Texas Instruments Incorporated
Most cited
By later US patents
- Senone tree representation and evaluation39US 5,794,197 AMicrsoft Corporation · 1998
- Method and apparatus for speech recognition using optimized partial mixture tying of HMM state functions37US 5,825,978 ASRI International · 1998
- Method of generating a subword model for speech recognition25US 5,677,988 AATR Interpreting Telephony Research Laboratories · 1997
- Discriminatively trained mixture models in continuous speech recognition19US 6,490,555 B1ScanSoft, Inc. · 2002
- Out of vocabulary pattern learning11US 9,607,618 B2Nice Systems Ltd. · 2017