PTAB
IPR2025-01067
Amazon.com Inc v. SoundClear Technologies LLC
Key Events
Petition
Table of Contents
petition
1. Case Identification
- Case #: IPR2025-01067
- Patent #: 9,070,374
- Filed: May 28, 2025
- Petitioner(s): Amazon.com, Inc., Amazon.com Services LLC, and Amazon Web Services, Inc.
- Patent Owner(s): SoundClear Technologies LLC
- Challenged Claims: 1-15
2. Patent Overview
- Title: Communication Apparatus and Condition Notification Method
- Brief Description: The ’374 patent describes a communication apparatus, such as a two-way radio, that includes a system to evaluate the quality of a user's speech. The apparatus uses this evaluation to control a light-emitting device (LED) to notify the user about the "sound pick-up state," such as whether transmitted speech is of good or bad quality.
3. Grounds for Unpatentability
Ground 1: Obviousness over Core Audio Feedback References - Claims 1-3, 5, 8-10, 13, and 15 are obvious over Yeager and Boillot.
- Prior Art Relied Upon: Yeager (Patent 7,574,361) and Boillot (Application # 2007/0129022).
- Core Argument for this Ground:
- Prior Art Mapping: Petitioner argued that Yeager disclosed the foundational communication apparatus—a two-way push-to-talk radio with a microphone, transmitter, and an LED indicator to show its communication mode (e.g., on during transmit, off during receive). Boillot was argued to disclose the missing inventive concept: a device that performs speech detection and speech quality evaluation (e.g., determining if a user's voice is sufficient to overcome ambient noise). Boillot’s system provides feedback to the user via an LED, which can blink or flash to indicate a "voicing quality problem." The combination of Yeager's radio with Boillot's speech quality feedback system was alleged to teach the key limitations of independent claims 1 and 9.
- Motivation to Combine: A POSITA would combine Boillot's speech quality feedback with Yeager’s radio to solve a problem Yeager itself identified: a user has no way of knowing if their speech is being transmitted properly. Adding Boillot's quality indicator would provide more granular and useful feedback (e.g., poor quality due to insufficient volume) beyond the simple transmit-status light taught by Yeager, representing a simple addition of a known element to obtain predictable results.
- Expectation of Success: A POSITA would have reasonably expected success, as the combination involved integrating a known type of signal processing (speech quality evaluation) into a conventional radio architecture to control a standard component (an LED), which would have been a trivial implementation.
Ground 2: Obviousness with Noise Cancellation - Claims 4, 7, 11, and 14 are obvious over Yeager, Boillot, and Chen.
- Prior Art Relied Upon: Yeager (Patent 7,574,361), Boillot (Application # 2007/0129022), and Chen (Application # 2009/0209290).
- Core Argument for this Ground:
- Prior Art Mapping: This ground built upon the Yeager and Boillot combination by adding Chen, which disclosed using a second microphone for active noise cancellation. Chen taught a communication device with a main microphone on the front (to capture the user's voice) and a sub-microphone on the back (to capture ambient noise). A noise cancellation module then subtracted the ambient noise from the user's speech signal to produce a clearer output for transmission. Petitioner argued this combination taught the dependent claims requiring a second sound pick-up unit and noise cancellation.
- Motivation to Combine: A POSITA would be motivated to add Chen's well-known dual-microphone noise cancellation technique to the Yeager/Boillot device to improve the quality of the transmitted speech, a primary goal of such systems. Yeager and Boillot both expressed a desire for improved audio transmission, and reducing background noise is a direct way to achieve that. It would have been logical to evaluate speech quality after the noise cancellation step to determine the quality of the final transmitted signal.
- Expectation of Success: The use of multiple microphones for noise cancellation was a widely known and conventional technique, making its addition to the base combination straightforward and predictable.
Ground 3: Obviousness with Spectral Analysis - Claims 8 and 15 are obvious over Yeager, Boillot, and Vähätalo.
Prior Art Relied Upon: Yeager (Patent 7,574,361), Boillot (Application # 2007/0129022), and Vähätalo (Patent 5,963,901).
Core Argument for this Ground:
- Prior Art Mapping: This ground addressed the specific limitations of claims 8 and 15, which required the speech-segment determination unit to convert an input signal into the frequency domain and determine if it contained a voice or noise component based on the resulting spectrum. Petitioner contended that Vähätalo disclosed this technique. Vähätalo taught a Voice Activity Detection (VAD) system that calculates power spectrum components of an input signal frame to make a more accurate and reliable determination of whether speech is present.
- Motivation to Combine: A POSITA would combine Vähätalo's more sophisticated, frequency-based VAD with the Yeager/Boillot device to improve the accuracy of the speech detection function. Improving the underlying VAD algorithm was a known path to enhancing overall system performance (e.g., more accurately switching between speech and comfort noise generation). This was presented as an obvious substitution of one known VAD technique for another to improve results.
- Expectation of Success: Implementing different known VAD algorithms was a matter of routine design choice for a POSITA, who would have reasonably expected that a more advanced spectral analysis technique like Vähätalo's would yield improved performance.
Additional Grounds: Petitioner asserted additional obviousness challenges against claims 5, 6, and 12 based on combinations including Visser (Application # 2011/0264447). Visser was cited for its teaching of using a dual-microphone setup to determine the direction of the main microphone based on the phase difference between the audio signals, and using that directional information to evaluate speech quality.
4. Relief Requested
- Petitioner requests institution of an inter partes review and cancellation of claims 1-15 of Patent 9,070,374 as unpatentable under 35 U.S.C. §103.
Analysis metadata