Speech and Audio Foundation Model Patents: Who Leads 2026
Speech and audio foundation model patents: who is filing, where claims concentrate, and what is still open
23.3% of all 1,636 records sit with just five assignees, but the next 90+ ranked companies still hold a meaningful long tail — this is a concentrated field, not a closed one. Annual filings rose from 84 in 2017 to a peak of 140 in 2024, with the 2021-to-2024 span alone up 103%. Because publication trails filing by roughly 18 months, the 2025 and 2026 counts (11 by mid-2026) are still filling in and should not be read as a slowdown.
- 1GOOGLE LLC8.1%
- 2MICROSOFT TECHNOLOGY LICENSING LLC7.3%
- 3SAMSUNG ELECTRONICS CO LTD2.9%
- 4INTERNATIONAL BUSINESS MACHINE CORPORATION2.5%
- 5NUANCE COMMUNICATIONS INC2.5%
See the full speech and audio foundation models analysis in Eureka
- The complete ranking, not just the top five
- Every IPC branch with its share of the corpus
- The most-cited records, and where claim space is still thin
Common questions on this landscape
Who holds the most patents in speech and audio foundation models?
One assignee leads the ranked list with 132 records, well ahead of fifth place at 41 and tenth place at 30. The top five assignees combined hold 23.3% of all 1,636 records in scope, and the top ten hold 32.6%, which means over two-thirds of the corpus is spread across a long tail of smaller and single-filing entrants. The ranking covers 100 companies total, so it should be read as the full returned list rather than a top-50 cut.
Is patent filing in this field still growing?
Filings rose from 69 in 2021 to a peak of 140 in 2024, a 103% increase over that span, and 2024 is the most recent year with substantially complete published data. Counts for 2025 and 2026 look lower, but that reflects the roughly 18-month lag between filing and publication rather than a genuine slowdown. Any claim about the field cooling off should be checked against complete years only.
What technology areas dominate the claim language?
G10L, the speech and audio analysis/synthesis subclass, appears in 79.5% of all 1,636 records, making it by far the largest single category. G06F (electric digital data processing) and G06N (AI-model computing) follow at 20.8% and 11.6% respectively, with smaller shares in image/video recognition, telephony and healthcare informatics. Because a single record can carry multiple IPC classes, these shares overlap and add up to more than 100%.
Disclaimer. This analysis is based on Patsnap Eureka data drawn from a limited snapshot of global patent records and is provided for general information and reference only. Patent data carries inherent limitations — recent filings are under-counted because of publication lag, counts may be on a record or family basis, classification and applicant-name data may contain errors or duplicates, and the underlying search query defines the scope shown — so the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing here is an exhaustive prior-art, novelty, freedom-to-operate or validity search, nor does it constitute legal, financial or professional advice, and it should not be relied upon as such. Verify independently and review with qualified patent and legal professionals before acting on it.
Method: Filing trend and technology composition. Derived from a Patsnap search on Speech and Audio Foundation Models covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish. Every share divides by all records in scope. Data: Patsnap Eureka. See the full landscape report.