PTAB

IPR2022-00005

Google LLC v. Vocalife LLC

Log In
Key Events
Petition

1. Case Identification

2. Patent Overview

  • Title: Microphone Array System
  • Brief Description: The ’371 patent describes methods and systems for enhancing a target sound signal, such as speech, by isolating it from ambient noise or other sounds. The disclosed system uses an array of microphones and several processing units: a sound source localization unit, an adaptive beamforming unit, a noise reduction unit, and an echo cancellation unit, which are implemented on a digital signal processor (DSP).

3. Grounds for Unpatentability

Ground 1: Claims 22, 23, 30, 31, and 38-41 are obvious over Mao in view of Briere, and optionally Van Trees

  • Prior Art Relied Upon: Mao (Application # 2005/0047611), Briere (a 2006 symposium paper), and Van Trees (a 2002 textbook).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Mao, a reference for reducing noise in audio signals, taught a microphone array system that included all the core functional units required by independent claims 22 and 30: a sound source localization unit, an adaptive beamforming unit, a noise reduction unit, and an echo cancellation unit. Mao disclosed using these components to locate, track, and enhance a target audio signal. Briere, which disclosed a sound localization and tracking system for a mobile robot, explicitly taught implementing these same functional modules on a single DSP to achieve real-time processing in a portable device.
    • Motivation to Combine: A POSITA would combine Mao and Briere to gain the known benefits of using a DSP architecture, such as improved processing speed, real-time performance, and power efficiency for a portable device, as described in Mao. Petitioner contended that Briere explicitly demonstrated the feasibility and advantages of integrating these known audio processing modules onto a single DSP. For limitations regarding specific time-delay-of-arrival (TDOA) calculations in non-linear arrays, Van Trees was cited as a foundational textbook providing the exact mathematical formulas, which a POSITA would have naturally consulted to implement Mao’s system in various microphone configurations.
    • Expectation of Success: The combination was argued to be a straightforward integration of known components for their intended purposes, yielding predictable results. Since both Mao and Briere described similar systems with similar components, implementing Mao’s system on a DSP as taught by Briere would have been a simple and predictable design choice.

Ground 2: Claims 22, 30, and 38-41 are obvious over Jeong, Van Trees, and Mao, and optionally Briere

  • Prior Art Relied Upon: Jeong (Korean Application # 10-2009-0037845), Van Trees, and Mao.
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner asserted that Jeong disclosed a method for extracting a target sound signal using a microphone array, a sound source searching unit (localization), an emphasized signal beam-former (adaptive beamforming), and a signal extracting unit (noise reduction). Petitioner argued that Jeong’s system was adaptive and capable of steering its directional response toward a moving sound source. To meet the echo cancellation limitation, Petitioner relied on Mao, which taught an acoustic echo cancellation (AEC) module. To provide the specific TDOA calculations that Jeong described at a high level, Petitioner again relied on the textbook teachings of Van Trees.
    • Motivation to Combine: A POSITA would have been motivated to modify Jeong’s apparatus to include Mao’s AEC module because both systems were designed for portable consumer devices where echoes are a known problem. Mao provided a known solution to this problem. Furthermore, because Jeong stated that its TDOA methods would be "readily understood" by a POSITA, one would have been motivated to consult a standard reference like Van Trees to implement the specific, well-known mathematical formulas for any given microphone array geometry.
    • Expectation of Success: Combining these references involved applying known solutions (Mao’s echo cancellation, Van Trees’s TDOA math) to a known system (Jeong’s beamformer) to achieve the predictable functions of echo removal and precise sound source localization.

Ground 3: Claims 22, 24-26, 28-30, 32-34, 36, and 38 are obvious over Jeong, Van Trees, and Mao, in view of Buck

  • Prior Art Relied Upon: Jeong, Van Trees, Mao, and Buck (Application # 2008/0144848).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground presented an alternative interpretation where the "adaptive beamforming unit" in Jeong comprised a combination of its emphasized beam-former, suppressed beam-former, and signal extracting unit. Under this view, Jeong lacked a separate, subsequent noise reduction unit. Petitioner argued that Buck, which disclosed an echo compensation system, taught a beam-formed filter circuit that functions as a post-filter to suppress residual echo and noise, thereby enhancing the quality of the beam-formed signal. This circuit from Buck was mapped to the "noise reduction unit" limitation.
    • Motivation to Combine: A POSITA would have been motivated to add Buck’s post-filter to Jeong’s system to improve sound quality, a common goal in audio processing. Petitioner argued that both Jeong and Buck utilized similar underlying components for beamforming (e.g., blocking matrices, adaptive filters), making the integration straightforward. Buck provided a known method for enhancing the output of a beamforming system like Jeong’s.
    • Expectation of Success: Implementing Buck's well-known post-filtering technique after Jeong's beamforming process would predictably result in reduced residual noise and enhanced signal quality without requiring substantial modification to Jeong's system.
  • Additional Grounds: Petitioner asserted additional obviousness challenges based on combinations including Yen (Application # 2009/0271187) for teaching adaptive step-size adjustment in filtering (for claims 27 and 35), and Andrea (Application # 2009/0268931) for teaching the use of A/D and D/A codecs (for claim 37).

4. Key Claim Construction Positions

  • Petitioner argued that no specific claim terms required construction and that the claims should be given their plain and ordinary meaning. This position was asserted to be consistent with the Patent Owner's own proposals in related district court litigation. Petitioner noted that even under a prior court construction of "adaptive beamforming" from a related patent, the cited prior art disclosed the limitation. The petition proceeded by analyzing the challenged claims under their plain and ordinary meaning, submitting that the Board need not undertake a means-plus-function analysis under §112, ¶ 6.

5. Arguments Regarding Discretionary Denial

  • Petitioner argued that discretionary denial under Fintiv would be improper. It asserted that the parallel district court litigation was in its early stages, with minimal investment from the parties and court, and that a stay pending the IPR was likely to be granted. Petitioner stipulated that, if the IPR was instituted, it would not pursue the same invalidity grounds in the district court, thus avoiding duplicative efforts. It was also argued that the merits of the petition were strong and based on prior art not considered during prosecution, favoring institution. Petitioner also contended that denial under the General Plastic factors was inappropriate.

6. Relief Requested

  • Petitioner requested institution of an inter partes review of claims 22-41 of Patent RE48,371 and cancellation of those claims as unpatentable.