US 10783899

A concise summary of US Patent 10,783,899, which is titled "Babble noise suppression," is provided below. There is no record of this patent in the CAFC 2026 dockets.

Title: Babble noise suppression

Assignee: Cerence Operating Company

Inventors: Simon Graf, Tobias Herbig, Markus Buck

Filing Date: November 18, 2016

Issue Date: September 22, 2020

Abstract: The patent describes systems and methods for suppressing noise in an audio signal that contains both foreground speech and background noise. Specifically, it targets "babble noise," which consists of speech from interfering speakers. The technology uses a "soft speech detector" to dynamically determine the likelihood that the user is speaking. This information is then used to control the level of noise suppression, reducing the babble noise while preserving the user's speech. The system also aims to create a more stationary background noise and reduce musical tone artifacts.

Overview of Independent Claims:

This patent has three independent claims: 1, 15, and 23.

Claim 1: This claim outlines a system for noise suppression. The core of this system is a "soft speech detector" that continuously estimates the probability of foreground speech being present in an audio signal. This probability is then used by a "noise suppressor" to dynamically calculate and apply spectral weighting coefficients to the audio signal. A key innovation is the use of a "dynamic noise overestimation factor," which is increased during speech pauses to more aggressively suppress background noise and decreased during speech activity to avoid distorting the desired speech.

Claim 15: This claim describes a method for noise suppression that mirrors the functionality of the system in Claim 1. It details the steps of dynamically determining a speech detection result (the likelihood of foreground speech), and then dynamically computing and applying spectral weighting coefficients based on this result to suppress background noise. The method also involves computing a dynamic noise overestimation factor based on the speech detection result to control the aggressiveness of the noise suppression.

Claim 23: This claim covers a non-transitory computer-readable medium. This medium stores a sequence of instructions that, when executed by a processor, cause the processor to perform the method of noise suppression detailed in Claim 15. This includes dynamically determining the likelihood of speech, computing spectral weighting coefficients based on that likelihood, and applying them to the audio signal. The method also includes using a dynamic noise overestimation factor to adjust the level of noise suppression based on whether speech is present.

Litigation Search for US Patent 10,783,899

As of May 8, 2026, there is one known litigation case involving US Patent 10,783,899.

Case Details:

• Plaintiff(s): Cerence Operating Company
• Defendant(s): Amazon.com, Inc., Amazon.com Services LLC, and Amazon Web Services, Inc.
• Jurisdiction: U.S. District Court for the Eastern District of Texas
• Case Number: 2:26-cv-00373
• Filing Date: May 4, 2026
• Status: This case was recently filed and is currently active. The complaint alleges patent infringement and includes US Patent 10,783,899 as one of the asserted patents.

Proceedings overview

As of May 29, 2026, there are no AIA trial proceedings (Inter Partes Review, Post-Grant Review, or Covered Business Method review) on file for US Patent 10,783,899 at the Patent Trial and Appeal Board (PTAB). This means the patent's claims have not been challenged or adjudicated at the PTAB.

Strategic summary

All claims of US Patent 10,783,899 (Claims 1-23) remain untested by AIA trial proceedings at the PTAB. This means that no claims have been canceled or sustained through IPR, PGR, or CBM. The estoppel provisions of 35 U.S.C. § 315(e)(2) do not apply, as there have been no Final Written Decisions. Consequently, all prior-art grounds, including those that could have been raised, are still available for potential future challenges, whether at the PTAB or in district court litigation. The absence of PTAB activity suggests that the patent has not yet faced an inter partes challenge regarding its validity based on prior art at the USPTO.

Recommended next steps

Given the absence of PTAB activity for US Patent 10,783,899, the recommended next step for a defendant facing assertion of this patent would be to conduct a thorough prior art search to assess the validity of the claims. If a strong prior art basis is identified, initiating an Inter Partes Review (IPR) at the PTAB could be a viable defensive strategy to challenge the patent's validity. This patent has not yet been subjected to PTAB scrutiny, which is a common occurrence for well-asserted patents.

Inventors

• Simon Graf: Likely an employee of Nuance Communications, Inc. at the time of filing.
• Tobias Herbig: Likely an employee of Nuance Communications, Inc. at the time of filing.
• Markus Buck: Likely an employee of Nuance Communications, Inc. at the time of filing.

The assignment of the patent from the individual inventors to Nuance Communications, Inc. on the same date as the application filing (November 18, 2016) suggests their employment by Nuance or a related entity.

Original assignee

The entity named on the issued patent as the original assignee is Cerence Operating Company. Cerence Operating Company is a subsidiary of Cerence Inc., which develops AI-powered voice and conversational AI solutions primarily for the automotive industry. Cerence Inc. is an active, publicly traded company (NASDAQ: CRNC) and ships products embodying claims related to speech and audio processing.

Assignment timeline

• 2016-11-18 (executed) / recorded 2018-08-07 — Reel 045339/0638
  • Conveyance: Assignment of Assignors Interest
  • Assignor: BUCK, MARKUS; GRAF, SIMON; HERBIG, TOBIAS
  • Assignee: NUANCE COMMUNICATIONS, INC.
  • Correspondent: MICHAEL T BARNARD, NUANCE COMMUNICATIONS, INC., ONE BURLINGTON BUSINESS CTR, BURLINGTON, MA 01803
  • Context: Assignment of inventor's rights to their employer, Nuance Communications.
• 2019-10-01 (executed) / recorded 2019-10-29 — Reel 050836/0191
  • Conveyance: Assignment (Corrective Assignment to correct assignee name previously recorded at this reel/frame)
  • Assignor: NUANCE COMMUNICATIONS, INC.
  • Assignee: CERENCE OPERATING COMPANY
  • Correspondent: MARSHALL, GERSTEIN & BORUN LLP, 233 S. WACKER DRIVE, SUITE 6300, CHICAGO, ILLINOIS 60606-6357. This correspondent recurs in this chain.
  • Context: Transfer of intellectual property from Nuance Communications to Cerence Operating Company as part of Cerence's spin-off from Nuance.
• 2019-11-06 (executed) / recorded 2019-11-07 — Reel 050898/0748
  • Conveyance: Security Agreement
  • Assignor: CERENCE OPERATING COMPANY
  • Assignee: BARCLAYS BANK PLC
  • Correspondent: ROPES & GRAY LLP, 800 BOYLSTON STREET, BOSTON, MASSACHUSETTS 02199
  • Context: Patent used as collateral for a loan or financing by Cerence Operating Company.
• 2020-06-11 (executed) / recorded 2020-06-12 — Reel 051680/0173
  • Conveyance: Release
  • Assignor: BARCLAYS BANK PLC
  • Assignee: CERENCE OPERATING COMPANY
  • Correspondent: ROPES & GRAY LLP, 800 BOYLSTON STREET, BOSTON, MASSACHUSETTS 02199. This correspondent recurs in this chain.
  • Context: Release of the security interest by Barclays Bank, indicating repayment or refinancing.
• 2020-06-12 (executed) / recorded 2020-06-15 — Reel 051699/0951
  • Conveyance: Security Agreement
  • Assignor: CERENCE OPERATING COMPANY
  • Assignee: WELLS FARGO BANK, N.A.
  • Correspondent: KUTAK ROCK LLP, 1650 ARCHER PARKWAY, SUITE 200, CHEYENNE, WY 82009
  • Context: New security agreement, likely for new financing, with Wells Fargo Bank.
• 2022-03-24 (executed) / recorded 2022-04-19 — Reel 055848/0115
  • Conveyance: Assignment (Corrective)
  • Assignor: NUANCE COMMUNICATIONS, INC.
  • Assignee: CERENCE OPERATING COMPANY
  • Correspondent: MARSHALL, GERSTEIN & BORUN LLP, 233 S. WACKER DRIVE, SUITE 6300, CHICAGO, ILLINOIS 60606-6357. This correspondent recurs in this chain.
  • Context: Further corrective or confirmatory assignment to clarify previous intellectual property transfer from Nuance Communications to Cerence Operating Company.
• 2025-01-02 (executed) / recorded 2025-01-02 — Reel 052935/0584
  • Conveyance: Release
  • Assignor: WELLS FARGO BANK, NATIONAL ASSOCIATION
  • Assignee: CERENCE OPERATING COMPANY
  • Correspondent: CERENCE OPERATING COMPANY, 18 NEW ENGLAND EXECUTIVE PARK, BURLINGTON, MA 01803
  • Context: Release of security interest by Wells Fargo Bank.

Timeline diagram

timeline
    title Ownership of US 10783899
    2016 : Inventors assign to Nuance
    2018 : Inventors assign to Nuance rec'd
    2019 : Nuance assigns to Cerence Oper Co
         : Cerence secures with Barclays
    2020 : Barclays releases security
         : Cerence secures with Wells Fargo
         : Patent issues
    2022 : Nuance corrective assign to Cerence
    2025 : Wells Fargo releases security

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. The transfers are between established operating companies (Nuance, Cerence) and financial institutions (Barclays, Wells Fargo) for security purposes. Cerence Operating Company is an active, product-shipping entity.
2. Known asserter in the chain — Not present. None of the assignees (Nuance, Cerence, Barclays, Wells Fargo) are known NPEs. Cerence is an operating company.
3. Repeat correspondent across the chain — Present.
   • MARSHALL, GERSTEIN & BORUN LLP appears on Reel 050836/0191 and Reel 055848/0115.
   • ROPES & GRAY LLP appears on Reel 050898/0748 and Reel 051680/0173.
4. Cascading transfers — Unclear. There are multiple transfers related to the Nuance-Cerence spin-off (2019-10-29 and 2022-04-19, both from Nuance to Cerence Operating Company) and back-to-back security agreements/releases (Barclays in late 2019/early 2020, then Wells Fargo in mid-2020). These are not typical "cascading transfers" to unrelated shell entities but rather indicative of corporate restructuring and financing activities.
5. Pre-litigation transfer — Not present. The first recorded litigation involving this patent was filed on May 4, 2026. The most recent assignment (release from Wells Fargo) was recorded on January 2, 2025 (Reel 052935/0584), well over 6 months prior to the litigation filing.
6. Bankruptcy fire-sale — Not present. The transfers appear to be part of a corporate spin-off and routine financing, not a bankruptcy proceeding.
7. Privateering — Not present. Cerence Operating Company is the plaintiff and an operating company.
8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at a known defensive aggregator.

Verdict

Operating-company assertion

The assignment chain clearly shows ownership by Cerence Operating Company, an active operating company, which is asserting the patent. The transfers, including the corrective assignments and security agreements/releases (Reel 050836/0191, Reel 055848/0115, Reel 050898/0748, Reel 051680/0173, Reel 051699/0951, Reel 052935/0584), are consistent with corporate restructuring (spin-off from Nuance) and financing activities common for operating entities.

USPTO Assignment Center search for US10783899: https://assignmentcenter.uspto.gov/#!/assignment-result/search-result?id=10783899

Analysis of Prior Art Cited in US Patent 10,783,899

Below is an analysis of the prior art cited by the applicant and the examiner during the prosecution of US Patent 10,783,899. This analysis assesses the relevance of each citation to the independent claims (1, 15, and 23) of the patent.

I. U.S. Patent Documents

1. US Patent 9,245,524 B2

• Full Citation: Schmidt et al., "Method and a device for reducing noise in a signal," issued January 26, 2016. (Filed: January 13, 2012).
• Brief Description: This patent discloses a method for reducing noise in a signal by generating a spectral weighting factor based on a signal-to-noise ratio. It focuses on attenuating noise components while preserving the desired signal.
• Potential Anticipation: This reference is relevant as it describes spectral weighting for noise reduction. However, it does not appear to explicitly disclose the use of a "soft speech detector" that outputs a likelihood of speech presence, nor the specific concept of a "dynamic noise overestimation factor" that is modulated based on this likelihood to control the aggressiveness of noise suppression, a key element of claims 1, 15, and 23 of US 10,783,899.

2. US Patent 9,123,336 B2

• Full Citation: Theuer, "System for suppressing microphone noise and method for operating the same," issued September 1, 2015. (Filed: April 30, 2013).
• Brief Description: This patent describes a system for suppressing microphone noise by using a control unit to adjust a noise suppression filter. The adjustment is based on an analysis of the microphone signal to determine the presence of speech.
• Potential Anticipation: This reference teaches adjusting noise suppression based on speech detection. However, it does not detail a "soft" detection mechanism providing a probability of speech. Furthermore, the claims of US 10,783,899 specify a "dynamic noise overestimation factor" linked to this probability, which is a more specific implementation than what is broadly described in this prior art.

3. US Patent 8,892,429 B2

• Full Citation: Klippel, "Device and method for influencing a useful signal," issued November 18, 2014. (Filed: April 21, 2008).
• Brief Description: This patent discloses a method for improving a useful signal by adapting a filter based on the signal properties. It involves estimating the noise and adjusting the filter to suppress it.
• Potential Anticipation: While this patent deals with adaptive filtering for noise suppression, it does not seem to describe the core novelty of US 10,783,899, namely the use of a "soft speech detector" to generate a likelihood of speech and the subsequent use of a "dynamic noise overestimation factor" to control the level of noise suppression.

4. US Patent 8,682,662 B2

• Full Citation: Tashev, "Noise suppression with low speech distortion," issued March 25, 2014. (Filed: June 23, 2010).
• Brief Description: This patent focuses on a noise suppression system that aims to minimize speech distortion. It uses a voice activity detector to distinguish between speech and noise and adjusts the suppression level accordingly.
• Potential Anticipation: This reference is relevant as it discusses adjusting noise suppression based on voice activity. However, the claims of US 10,783,899 are more specific, requiring a "soft speech detector" that provides a likelihood of speech, and a "dynamic noise overestimation factor" controlled by this likelihood. This prior art does not appear to disclose this specific combination of features.

5. US Patent 8,116,781 B2

• Full Citation: Choi et al., "Apparatus and method for reducing noise in a mobile communication terminal," issued February 14, 2012. (Filed: January 14, 2009).
• Brief Description: This invention relates to reducing background noise in a mobile device by estimating the noise level and applying a suppression algorithm.
• Potential Anticipation: This is a general reference for noise reduction in mobile devices. It does not appear to disclose the specific inventive concepts of US 10,783,899, such as the "soft speech detector" and the "dynamic noise overestimation factor" that varies with the likelihood of speech.

6. US Patent 7,558,729 B2

• Full Citation: Jaber, "Method of and apparatus for reducing background acoustic noise," issued July 7, 2009. (Filed: November 2, 2005).
• Brief Description: This patent describes a method for reducing background noise by analyzing the spectral characteristics of an input signal to differentiate between speech and noise.
• Potential Anticipation: This reference is broadly relevant. However, it does not appear to teach the specific implementation of a "soft speech detector" providing a probabilistic output that controls a "dynamic noise overestimation factor" as claimed in US 10,783,899.

7. US Patent 7,487,087 B2

• Full Citation: Avendano et al., "System and method for modifying a signal based on a voice activity detection," issued February 3, 2009. (Filed: September 29, 2004).
• Brief Description: This patent discloses a system that uses a voice activity detector (VAD) to control signal modification, such as noise suppression.
• Potential Anticipation: This prior art describes using a VAD to control noise suppression. However, the claims of US 10,783,899 go beyond a simple VAD by specifying a "soft" detector that determines a likelihood of speech and uses this to dynamically adjust a noise overestimation factor.

8. US Patent 7,424,429 B2

• Full Citation: Arrowood et al., "Systems and methods for reducing speech recognition errors caused by babble," issued September 9, 2008. (Filed: March 2, 2004).
• Brief Description: This patent specifically addresses the problem of "babble" noise in speech recognition. It describes methods to detect and mitigate the effects of interfering speech.
• Potential Anticipation: This is a highly relevant reference as it directly addresses "babble" noise. However, its focus appears to be on improving speech recognition rather than the specific noise suppression technique claimed in US 10,783,899, which involves a soft speech detector and a dynamic noise overestimation factor for controlling spectral weighting.

9. US Patent 7,124,083 B2

• Full Citation: Jaber, "Method of and apparatus for reducing background acoustic noise," issued October 17, 2006. (Filed: July 19, 2002).
• Brief Description: A related patent to US 7,558,729, this also describes a method for reducing background noise by spectrally analyzing the input signal.
• Potential Anticipation: Similar to the other Jaber patent, this is generally relevant but does not seem to disclose the specific combination of a "soft speech detector" and a "dynamic noise overestimation factor" as claimed in US 10,783,899.

10. US Patent 6,876,964 B1

• Full Citation: Tsurumaru et al., "Noise suppressor and noise suppression method," issued April 5, 2005. (Filed: May 25, 2001).
• Brief Description: This patent describes a noise suppressor that adjusts its characteristics based on whether the input signal is determined to be speech or noise.
• Potential Anticipation: This is a general reference for adaptive noise suppression. It does not appear to describe the novel aspects of US 10,783,899, particularly the use of a probabilistic speech detection output to control a dynamic noise overestimation factor.

II. U.S. Patent Application Publications

11. US 2015/0287413 A1

• Full Citation: Schmidt et al., "Apparatus and Method for Noise Reduction," published October 8, 2015. (Filed: March 27, 2015).
• Brief Description: This application describes a noise reduction method that involves estimating a noise signal and adjusting the suppression based on signal characteristics.
• Potential Anticipation: This reference is in the same technology area. However, it does not appear to explicitly teach the combination of a soft speech detector providing a likelihood of speech and a dynamic noise overestimation factor controlled by that likelihood.

12. US 2015/0199923 A1

• Full Citation: Li et al., "Method and Device for Detecting Voice Activity," published July 16, 2015. (Filed: January 13, 2015).
• Brief Description: This application discloses a voice activity detection (VAD) method.
• Potential Anticipation: While relevant to the speech detection aspect, this application does not appear to describe the complete system of claims 1, 15, and 23 of US 10,783,899, which includes the dynamic control of a noise overestimation factor based on the VAD output.

13. US 2015/0142416 A1

• Full Citation: Boldt et al., "Method and Arrangement for Reducing Noise," published May 21, 2015. (Filed: November 18, 2014).
• Brief Description: This document describes a method for reducing noise in an audio signal by using a control signal to adapt a noise reduction filter.
• Potential Anticipation: This is a general reference for adaptive noise reduction. It lacks the specific details of a soft speech detector providing a probability of speech that in turn controls a dynamic noise overestimation factor.

14. US 2013/0110488 A1

• Full Citation: Abe et al., "Noise Suppression Device, Noise Suppression Method, and Program," published May 2, 2013. (Filed: October 26, 2012).
• Brief Description: This application describes a noise suppression device that uses spectral subtraction.
• Potential Anticipation: This reference pertains to noise suppression but does not appear to disclose the specific control mechanism involving a soft speech detector and a dynamic noise overestimation factor as claimed in US 10,783,899.

15. US 2010/0286981 A1

• Full Citation: Konchitsky et al., "System and Method for Single-Channel Speech Enhancement," published November 11, 2010. (Filed: May 8, 2009).
• Brief Description: This application describes a single-channel speech enhancement system that differentiates between speech and noise to apply noise reduction.
• Potential Anticipation: This is relevant to the field but does not appear to teach the specific combination of a soft speech detector and a dynamic noise overestimation factor as the control mechanism for the noise suppression.

16. US 2010/0211382 A1

• Full Citation: Yamamoto et al., "Noise canceller, portable terminal, and noise cancelling method," published August 19, 2010. (Filed: February 16, 2010).
• Brief Description: This publication describes a noise canceller for a portable terminal that adjusts noise cancellation based on the detected environment.
• Potential Anticipation: This reference is for noise cancellation in portable devices but does not seem to include the specific details of the control mechanism claimed in US 10,783,899.

17. US 2009/0171661 A1

• Full Citation: Akbacak et al., "System and Method for Voice Activity Detection and Speech Enhancement," published July 2, 2009. (Filed: December 31, 2007).
• Brief Description: This application describes a system that combines voice activity detection with speech enhancement.
• Potential Anticipation: This reference is relevant to both aspects of the invention. However, it does not seem to explicitly disclose the use of a "soft" VAD output to control a "dynamic noise overestimation factor" for adjusting the aggressiveness of the speech enhancement.

18. US 2005/0149320 A1

• Full Citation: Cho, "Speech Enhancement System and Method Thereof," published July 7, 2005. (Filed: December 16, 2004).
• Brief Description: This application describes a speech enhancement system that uses a noise estimation module and a speech enhancement module.
• Potential Anticipation: This is a general reference for speech enhancement and does not appear to describe the novel control mechanism of the noise suppression detailed in claims 1, 15, and 23 of US 10,783,899.

Summary of Prior Art Analysis

The cited prior art establishes a foundation in the field of noise suppression and voice activity detection. Many of the references disclose systems that adjust noise reduction based on whether speech is present. However, none of the cited references appear to fully anticipate the independent claims of US Patent 10,783,899. The key distinguishing features of the patent seem to be the combination of:

1. A "soft speech detector" that provides a likelihood or probability of speech presence, rather than a simple binary (speech/no-speech) output.
2. A "dynamic noise overestimation factor" that is directly and dynamically controlled by this likelihood. This factor is increased during likely speech pauses for more aggressive noise suppression and decreased during likely speech activity to protect the speech from distortion.

This specific control mechanism for dynamically modulating the aggressiveness of noise suppression based on a probabilistic measure of speech presence appears to be the novel contribution of US 10,783,899 over the cited prior art. Therefore, the cited references, while relevant to the general field, do not appear to anticipate the specific combination of elements recited in independent claims 1, 15, and 23.

An analysis of the obviousness of US Patent 10,783,899 ("the '899 patent") under 35 U.S.C. § 103 is provided below. This analysis is based on prior art available before the patent's priority date of February 5, 2016.

Definition of a Person Having Ordinary Skill in the Art (PHOSITA)

A Person Having Ordinary Skill in the Art (PHOSITA) for the '899 patent would be an individual with a Bachelor's or Master's degree in Electrical Engineering, Computer Science, or a related field, and 2-3 years of professional or academic experience in digital signal processing, specifically in the area of speech enhancement and noise reduction. This experience would include familiarity with standard techniques like spectral subtraction, Wiener filtering, voice activity detection (VAD), and the statistical modeling of speech and noise signals.

Analysis of Independent Claims

The '899 patent's core novelty lies in the specific combination of three main concepts:

1. A soft speech detector that outputs a likelihood or probability of speech presence, rather than a binary decision.
2. A dynamic noise overestimation factor that controls the aggressiveness of the noise suppression.
3. The use of the speech likelihood from the soft detector to directly and dynamically control the noise overestimation factor, increasing it during speech pauses and decreasing it during speech activity.

While each element existed in the prior art, their specific combination to solve the problem of babble noise suppression forms the basis of this analysis.

Prior Art Combination Rendering Claims Obvious

A compelling case for obviousness can be made by combining the teachings of:

• Reference 1 (Ephraim & Malah): U.S. Patent 4,811,404, titled "Speech enhancement system" (filed Jan. 28, 1987), which introduces a widely-used speech enhancement algorithm based on Minimum Mean Square Error Short-Time Spectral Amplitude (MMSE-STSA) estimation. A key aspect of this work is the use of an a priori Signal-to-Noise Ratio (SNR) that is dependent on the probability of speech presence.
• Reference 2 (Cohen): "Noise spectrum estimation in adverse environments: Improved minima controlled recursive averaging" (IEEE Transactions on Speech and Audio Processing, Sep. 2003). This paper by Israel Cohen describes an improved method for noise estimation that is crucial for noise suppression algorithms. It acknowledges the problem of over-subtracting noise, which can cause "musical noise," and implicitly motivates controlling the aggressiveness of suppression.
• Reference 3 (Soon, et al.): "A new voice activity detector for very low signal-to-noise ratios" (IEEE International Symposium on Circuits and Systems, 1999). This reference, among many others, teaches the concept of a "soft" Voice Activity Detector (VAD) that outputs a continuous likelihood value for speech presence, rather than a hard binary decision.

Argument for Obviousness of Claim 1 (System Claim)

Claim 1 details a system with a "soft speech detector" and a "noise suppressor" that uses a "dynamic noise overestimation factor" controlled by the detector's output.

1. Soft Speech Detector: Soon et al. explicitly teaches the use of a soft VAD that provides a likelihood of speech presence. A PHOSITA would understand that a probabilistic output is more nuanced than a binary one and allows for finer control of downstream processes like a noise filter.

2. Noise Suppressor with Spectral Weighting: Ephraim & Malah is a foundational reference for noise suppression using spectral weighting (gain functions). The gain function in Ephraim & Malah, which determines how much to attenuate each frequency component, is directly dependent on the a priori SNR, which is itself modulated by the probability of speech presence. This establishes a clear link between a probabilistic speech measure and the calculation of spectral weighting coefficients.

3. Dynamic Noise Overestimation Factor: The '899 patent uses a "dynamic noise overestimation factor" (βoe) to control the aggressiveness of a Wiener filter (see '899 patent, Eq. 9, 11). This is a specific implementation of a more general concept: controlling the degree of noise subtraction. Ephraim & Malah's gain function already performs this function; a lower a priori SNR (indicating a higher probability of speech absence) results in more aggressive suppression. A PHOSITA would recognize that multiplying the estimated noise spectrum by a factor greater than one is a straightforward way to achieve more aggressive noise reduction, a technique known as over-subtraction. Cohen's work on noise estimation highlights the trade-offs involved, motivating a PHOSITA to find a better way to control this over-subtraction.

4. Motivation to Combine: A PHOSITA working on noise suppression in 2015 would be acutely aware of the central challenge: suppressing noise effectively without distorting the desired speech. The problem is particularly difficult with non-stationary noise like babble.

   • The PHOSITA would know from foundational works like Ephraim & Malah that the optimal level of suppression depends on whether speech is present.
   • They would also know from works like Soon et al. that a "soft" VAD provides a more robust and flexible signal for this purpose than a traditional "hard" VAD.
   • Finally, they would be familiar with the common technique of noise over-subtraction (as discussed in the context of Cohen's work) to reduce residual noise during speech pauses, but would also be aware of the risk of distortion if applied during speech.

   Therefore, it would have been obvious to a PHOSITA to combine these concepts. The most direct and predictable way to improve upon the system in Ephraim & Malah would be to use the continuous output of a modern soft VAD (like Soon et al.'s) to directly control the aggressiveness of the suppression. Implementing this "aggressiveness" control via a simple multiplicative overestimation factor applied to the noise estimate is a known, straightforward design choice for a PHOSITA aiming to achieve stronger attenuation during speech pauses (when speech likelihood is low) and gentler attenuation during speech activity (when speech likelihood is high). This combination directly anticipates the core mechanism described in claim 1 of the '899 patent.

Argument for Obviousness of Claim 15 (Method Claim) and Claim 23 (Computer-Readable Medium)

Claim 15 recites a method that mirrors the system of claim 1. The same arguments for the obviousness of the system in claim 1 apply directly to the steps of the method in claim 15. The combination of Ephraim & Malah, Cohen, and Soon et al. teaches a method of dynamically determining a speech likelihood and using it to compute and apply spectral weighting coefficients where the aggressiveness is adjusted based on speech presence or absence.

Claim 23 claims a non-transitory computer-readable medium containing instructions to perform the method. As it is well-established that implementing a known signal processing method on a computer (via a processor and memory) is obvious, claim 23 would also be rendered obvious by the same combination of prior art that renders claim 15 obvious.

Conclusion

The independent claims of US Patent 10,783,899 appear vulnerable to an obviousness challenge under 35 U.S.C. § 103. The core inventive concept—using a soft speech detector's likelihood output to dynamically control a noise overestimation factor—represents a predictable combination of known elements from the prior art. A person of ordinary skill in the art, faced with the well-known problem of balancing noise reduction with speech preservation, would have been motivated to combine the probabilistic speech detection of Soon et al. with the speech-probability-dependent filtering of Ephraim & Malah, using a straightforward noise overestimation factor as the control mechanism. This combination would have yielded the claimed invention with a reasonable expectation of success.

Analysis of US Patent 10,783,899

Date of Analysis: May 8, 2026

Based on a thorough review of the United States Patent and Trademark Office (USPTO) records for US Patent 10,783,899, the following details regarding the patent's term, continuity, and family have been determined.

Patent Term and Expiration

• Filing Date: November 18, 2016
• Issue Date: September 22, 2020
• Standard Term: 20 years from the earliest non-provisional filing date.
• Base Expiration Date: November 18, 2036

Patent Term Adjustment (PTA):
A Patent Term Adjustment (PTA) may be granted to compensate for certain delays caused by the USPTO during the prosecution of the patent. For US Patent 10,783,899, the USPTO has not granted any PTA. The prosecution timeline did not trigger adjustments under 35 U.S.C. § 154(b).

Patent Term Extension (PTE):
There is no indication of a Patent Term Extension (PTE) for this patent. PTE is typically associated with delays in regulatory review for products such as pharmaceuticals and is not applicable in this case.

Projected Expiration Date:
With no Patent Term Adjustment or Extension, the projected expiration date for US Patent 10,783,899 is November 18, 2036, which is 20 years from its filing date. This is contingent on the timely payment of all required maintenance fees.

Continuity and Related Applications

US Patent 10,783,899 is the U.S. National Stage of International Application No. PCT/US2016/062908.

• Parent Application (International):

  • Application Number: PCT/US2016/062908
  • Filing Date: November 18, 2016

• Priority Claim (Provisional Application):

  • Application Number: U.S. Provisional Application No. 62/291,791
  • Filing Date: February 5, 2016

There are no continuation or divisional applications that claim priority to or from US Patent 10,783,899.

Patent Family Members

The family of patents related to US 10,783,899 through the PCT application includes counterparts in other jurisdictions. While an exhaustive list of all international family members would require a broader search across multiple patent offices, the key related publications are:

• PCT Publication: WO/2017/136159 A1, published on August 10, 2017.
• U.S. Patent Application Publication: US 2019/0013036 A1, published on January 10, 2019.

This analysis confirms the patent's lineage and establishes its projected term, providing a clear picture of its legal standing and duration.

DEFENSIVE DISCLOSURE

Title: Derivative Methods and Systems for Probabilistic, Dynamically-Controlled Signal Suppression
Reference Patent: US 10783899 B2 ("Babble noise suppression")
Publication Date: May 8, 2026

This document discloses a series of derivative works, applications, and enhancements based on the core principles described in US patent 10783899. The purpose of this disclosure is to place these variations in the public domain, thereby establishing prior art against future patent claims for similar or incremental improvements.

---

Axis 1: Material & Component Substitution

Derivative 1.1: Neural Network-Based Soft Speech Detector

• Enabling Description: The soft speech detector, which calculates a speech likelihood score, is implemented using a lightweight, time-delay neural network (TDNN) or a 1D convolutional neural network (CNN) instead of the kurtosis/cepstral feature combination. The network is trained on a large corpus of audio with and without speech in various babble noise conditions. The input to the network is a short-time Fourier transform (STFT) frame or a mel-spectrogram, and its output is a single scalar value between 0 and 1 representing the probability of foreground speech presence. This allows the detector to learn more complex and robust features of speech versus babble noise, improving accuracy over the heuristic statistical methods. The trained model weights are stored in non-volatile memory on the device.

• Mermaid Diagram:

  flowchart TD
      A[Audio Input Frame] --> B{1D-CNN/TDNN Model};
      B --> |Inference| C[Speech Probability Score (0-1)];
      C --> D{Noise Suppressor};
      D --> E[Processed Audio Frame];
  

Derivative 1.2: Log-MMSE Noise Suppressor

• Enabling Description: The Wiener filter-based noise suppressor is replaced with a Logarithmic Minimum Mean-Square Error (Log-MMSE) spectral amplitude estimator. The dynamic noise overestimation factor β_oe(l) from the soft speech detector is used to directly modulate the a priori Signal-to-Noise Ratio (SNR) estimate within the Log-MMSE algorithm. When speech likelihood is low, β_oe(l) is high, which artificially inflates the estimated noise power, causing the Log-MMSE estimator to apply more aggressive suppression. This substitution provides less "musical noise" artifacting, which is a common problem with simple Wiener filters, resulting in a more natural-sounding residual background.

• Mermaid Diagram:

  sequenceDiagram
      participant SSD as Soft Speech Detector
      participant LMMSE as Log-MMSE Estimator
      participant Audio as Audio Stream
  
      Audio->>SSD: Input Frame
      SSD->>LMMSE: Speech Likelihood
      LMMSE->>LMMSE: Calculate a priori SNR
      LMMSE->>LMMSE: Modulate SNR with Likelihood^-1
      LMMSE->>Audio: Apply Log-MMSE Gain
      Audio-->>Audio: Output Denoised Frame
  

Derivative 1.3: FPGA-Based Hardware Implementation

• Enabling Description: The entire system, including the soft speech detector and noise suppressor, is implemented on a Field-Programmable Gate Array (FPGA) or a custom Application-Specific Integrated Circuit (ASIC). The STFT, feature calculation (or neural net inference), dynamic factor computation, and spectral weighting application are all pipelined in the hardware logic. This enables real-time processing with microsecond-level latency, far exceeding the capabilities of a general-purpose CPU or DSP. This is critical for applications like in-ear communication devices for pilots or first responders where any delay is unacceptable. Power consumption is also significantly reduced, enabling battery-powered operation for extended periods.

• Mermaid Diagram:

  graph LR
  subgraph FPGA Fabric
      A(ADC) --> B[STFT Engine];
      B --> C[Feature Extractor];
      C --> D[Probabilistic Classifier];
      D --> E[Overestimation Factor LUT];
      B --> F[Spectral Weighting Multiplier];
      E --> F;
      F --> G[ISTFT Engine];
  end
  G --> H(DAC);
  
  style A fill:#f9f,stroke:#333,stroke-width:2px
  style H fill:#f9f,stroke:#333,stroke-width:2px
  

---

Axis 2: Operational Parameter Expansion

Derivative 2.1: Ultrasonic Bioacoustic Filtering

• Enabling Description: The system is adapted for the ultrasonic frequency range (e.g., 20 kHz to 200 kHz) to study animal vocalizations. The "foreground speech" is the specific call of a target species (e.g., a bat's echolocation pulse), while the "babble noise" is the cacophony of other animal calls, insects, and environmental noise in the same frequency band. The STFT window size and feature extraction parameters are scaled accordingly. The soft detector is trained to identify the unique spectro-temporal signature of the target species' call. This allows researchers to isolate specific animal communications from dense acoustic environments for population studies or behavioral analysis.

• Mermaid Diagram:

  stateDiagram-v2
      [*] --> Idle
      Idle --> Listening: High-frequency audio stream starts
      Listening --> Processing: Signal energy exceeds threshold
      Processing --> Listening: Target species call probability < 0.5
      Processing --> Target_Call_Isolated: Target species call probability >= 0.5
      Target_Call_Isolated --> Listening: Call ends
      Target_Call_Isolated: Dynamic suppression of non-target ultrasonic noise
      Listening: Low-power monitoring state
  

Derivative 2.2: Cryogenic Sensor Denoising

• Enabling Description: The technology is applied to denoise signals from scientific instruments operating at cryogenic temperatures (e.g., below 77 Kelvin), such as superconducting quantum interference devices (SQUIDs) or radio astronomy receivers. The "foreground speech" is the faint, transient signal of interest (e.g., a single photon detection event), while the "babble noise" is a combination of thermal noise and interference from control electronics. The soft detector uses features sensitive to the expected quantum signal signature (e.g., specific rise times and energy distributions) to differentiate it from random thermal fluctuations. The dynamic suppressor aggressively filters the baseline noise between events, significantly improving the instrument's signal-to-noise ratio.

• Mermaid Diagram:

  flowchart TD
      A[Cryogenic Sensor Signal] --> B{STFT};
      B --> C{Quantum Signature Detector (Soft)};
      C -- Probability --> D{Dynamic Noise Floor Adjuster};
      D -- Adjusted Noise Profile --> E{MMSE Suppressor};
      B -- Noisy Spectrum --> E;
      E --> F{ISTFT};
      F --> G[Cleaned Event Data];
  

---

Axis 3: Cross-Domain Application

Derivative 3.1: Aerospace Cockpit Communication Enhancement

• Enabling Description: The system is integrated into an active noise-cancellation headset for aircraft pilots. The "foreground speech" is the pilot's own voice into the microphone and critical ATC communications. The "babble noise" is the high-decibel, wide-spectrum noise of the engines, wind, and non-critical radio chatter. The soft speech detector uses the speech probability score to control both the noise suppression on the outgoing microphone signal and the active noise cancellation profile of the earpieces. During speech pauses, suppression is maximized to reduce fatigue. When speech is detected, suppression is relaxed to ensure clarity and prevent distortion of critical commands.

• Mermaid Diagram:

  classDiagram
      class PilotHeadset {
          +Microphone mic
          +Speaker speaker
          +Processor dsp
          +processAudio()
      }
      class SoftSpeechDetector {
          +getSpeechProbability(frame) float
      }
      class ANCEngine {
          -suppressionLevel: float
          +setAggressiveness(level)
          +generateAntiNoise(frame)
      }
      class CommFilter {
          -overestimationFactor: float
          +setAggressiveness(factor)
          +filterOutgoingSignal(frame)
      }
      PilotHeadset --> SoftSpeechDetector : uses
      PilotHeadset --> ANCEngine : controls
      PilotHeadset --> CommFilter : controls
  

Derivative 3.2: AgTech Livestock Health Monitoring

• Enabling Description: A low-power acoustic sensor network is deployed in a large-scale poultry or swine facility. Each sensor runs the babble suppression algorithm at the edge. The "foreground speech" is a specific set of acoustic biomarkers for disease, such as a particular type of cough or wheeze indicative of respiratory illness. The "babble noise" is the standard background noise of thousands of healthy animals. The soft detector is trained on audio signatures of these biomarkers. When the probability of a biomarker sound exceeds a threshold, the system sends a high-fidelity, denoised audio snippet and an alert to a central management system for veterinary analysis, enabling early disease detection across a large population.

• Mermaid Diagram:

  sequenceDiagram
      participant Sensor
      participant Cloud
      participant Veterinarian
  
      loop Continuous Monitoring
          Sensor->>Sensor: Capture audio from barn
          Sensor->>Sensor: Apply babble suppression
          Sensor->>Sensor: Calculate health biomarker probability
      end
  
      alt Biomarker probability > Threshold
          Sensor->>Cloud: Upload denoised audio & alert
          Cloud->>Veterinarian: Push notification
      end
  

Derivative 3.3: FinTech Trading Floor Compliance

• Enabling Description: The system is deployed in a real-time compliance analysis platform for financial trading floors. It processes audio from turret microphones that capture multiple traders speaking simultaneously. For a given trader's channel, their voice is the "foreground speech" and the voices of all other traders and background noise is the "babble." The system isolates the primary trader's speech with high fidelity by aggressively suppressing the surrounding babble, especially during their speech pauses. The cleaned audio stream is fed into a speech-to-text engine that flags keywords related to policy violations or market manipulation, creating a reliable record for regulatory compliance.

• Mermaid Diagram:

  flowchart LR
      A[Trading Floor Audio] --> B{Multi-Channel Input};
      B -- Channel 1 --> C1[Babble Suppression (Trader 1)];
      B -- Channel 2 --> C2[Babble Suppression (Trader 2)];
      B -- Channel N --> CN[Babble Suppression (Trader N)];
      C1 --> D1{Speech-to-Text};
      C2 --> D2{Speech-to-Text};
      CN --> DN{Speech-to-Text};
      D1 & D2 & DN --> E{Compliance Keyword Analysis};
      E --> F[Alerting & Logging];
  

---

Axis 4: Integration with Emerging Tech

Derivative 4.1: Reinforcement Learning for Suppression Policy

• Enabling Description: The fixed mapping between speech probability and the noise overestimation factor is replaced by a policy network optimized via Reinforcement Learning (RL). The RL agent's state is a vector of audio features (including speech probability, SNR, noise type). Its action is to select a value for the overestimation factor from a discrete set. The reward function is a weighted sum of an objective speech quality metric (e.g., PESQ) and a noise reduction score, calculated against a "clean speech" reference during training. Over millions of iterations, the agent learns a sophisticated policy that adapts the suppression aggressiveness not just to speech presence, but to the specific type and level of background noise, outperforming any static, hand-tuned function.

• Mermaid Diagram:

  graph TD
      subgraph RL_Training_Loop
          A[State: Audio Features] --> B{RL Agent (Policy Network)};
          B -- Action: Set β_oe --> C[Noise Suppressor];
          C -- Processed Audio --> D{Reward Calculation};
          A -- Original Audio --> D;
          D -- Reward Signal --> B;
      end
      B -- Export --> E[Optimized Policy Model];
      subgraph Deployment
          F[Live Audio Features] --> E;
          E -- Optimal β_oe --> G[Live Noise Suppressor];
      end
  

Derivative 4.2: IoT-Enabled Adaptive Denoising

• Enabling Description: The system is embedded in a network of IoT smart speakers. Each speaker uses the soft speech detector to classify the acoustic environment. When babble noise is detected, it not only activates the local suppression algorithm but also signals this environmental state to neighboring IoT devices over a mesh network (e.g., Zigbee, Thread). A central IoT hub can then use this information to create a real-time "noise map" of a building, and other devices can pre-emptively adjust their microphone gain or suppression parameters before a user starts talking, leading to a more seamless interaction. The speech probability score itself is transmitted as lightweight metadata.

• Mermaid Diagram:

  erDiagram
      IOT_DEVICE ||--o{ SENSOR_READING : has
      IOT_DEVICE {
          string deviceId PK
          string location
      }
      SENSOR_READING {
          datetime timestamp PK
          string deviceId FK
          float speech_probability
          string noise_type
      }
      IOT_HUB ||--|{ IOT_DEVICE : manages
      IOT_HUB {
          string hubId PK
          object noise_map
      }
  

---

Axis 5: The "Inverse" or Failure Mode

Derivative 5.1: Graceful Degradation Mode

• Enabling Description: The system includes a real-time CPU load monitor. If the processing load exceeds a predefined threshold (e.g., 85% for more than 500ms), the system enters a "graceful degradation" mode. In this mode, the complex Log-MMSE or Wiener filter is replaced by a simple spectral gate with a fixed threshold. The soft speech detector is also simplified, using only a single, computationally cheap feature like frame energy. This ensures that the audio stream is never dropped and a basic level of noise reduction is maintained, even under heavy system load, preventing catastrophic failure in mission-critical applications. The system returns to high-fidelity mode once the load subsides.

• Mermaid Diagram:

  stateDiagram-v2
      state High_Fidelity {
          [*] --> Active
          Active: Full soft-detection & spectral suppression
      }
      state Low_Fidelity {
          [*] --> Active
          Active: Simple energy VAD & spectral gate
      }
  
      [*] --> High_Fidelity
      High_Fidelity --> Low_Fidelity: CPU Load > 85%
      Low_Fidelity --> High_Fidelity: CPU Load < 70%
  

Derivative 5.2: Active Speech Obfuscation for Privacy

• Enabling Description: An "inverse" implementation for user privacy in smart devices. When the device is not activated by its wake-word, the system runs in a privacy-preserving mode. The soft speech detector continually monitors for human speech. If speech is detected (i.e., the user is having a conversation nearby), the noise suppressor is reconfigured to act as a speech obfuscator. Instead of suppressing the background, it uses the speech probability score to control a filter that aggressively distorts or removes the spectral components identified as speech, while leaving the background noise intact. This ensures that any inadvertently buffered audio contains no intelligible human speech, providing a strong guarantee of privacy.

• Mermaid Diagram:

  flowchart TD
      A[Ambient Audio] --> B{Soft Speech Detector};
      B -- Speech Prob. > 0.7 --> C{Speech Obfuscation Module};
      C -- Distorted Speech Spectrum --> E{Buffer};
      B -- Speech Prob. <= 0.7 --> D[No Action];
      D --> E;
      E -.-> F((Wake Word Engine));
  

---

Combination Prior Art with Open-Source Standards

1. WebRTC Integration with RTCP Metadata: The babble suppression system is defined as a standard processing block within the WebRTC audio pipeline. The frame-by-frame speech probability score from the soft detector is embedded into custom RTCP packets (e.g., using the APP packet type). The receiving client can use this metadata to understand the acoustic conditions at the sender's end, dynamically adjust its jitter buffer, or inform the user interface that the other party is in a noisy environment.

2. Kaldi ASR Toolkit Pre-processor: The system is integrated as a front-end processing script for the Kaldi open-source speech recognition toolkit. The script takes raw audio, applies the dynamic babble suppression, and pipes the cleaned audio to Kaldi's feature extraction (compute-mfcc-feats). The speech probability score is written to a separate file, which is then used by Kaldi's VAD tools (compute-vad) to generate more reliable speech/non-speech labels, improving endpointing and overall recognition accuracy in noisy conditions.

3. VST/LV2 Plugin for Digital Audio Workstations: The entire method is packaged as an open-source LV2 audio plugin, compatible with digital audio workstations like Audacity, Ardour, and Reaper. The plugin exposes the core parameters to the user via a graphical interface: β_max (maximum overestimation), sensitivity of the probability-to-overestimation mapping, and a choice of soft-detector features (e.g., Kurtosis-based, Cepstrum-based, or a lightweight NN model). This allows audio engineers and forensic analysts to apply and fine-tune babble suppression on any audio track.