Samsung Electronics Co Ltd v. Cerence Operating Co

1. Case Identification

• Case #: IPR2025-00458
• Patent #: 8,682,671
• Filed: January 21, 2025
• Petitioner(s): Samsung Electronics Co., Ltd., and Samsung Electronics America, Inc.
• Patent Owner(s): Cerence Operating Company
• Challenged Claims: 1-20

2. Patent Overview

• Title: Generating Synthetic Speech with Contrastive Stress
• Brief Description: The ’671 patent relates to techniques for improving text-to-speech (TTS) technology by generating synthetic speech with contrastive stress. The system analyzes input text to identify differing portions between related text strings and applies acoustic emphasis (e.g., increased pitch, amplitude, or duration) to those differing portions to mimic natural human speech patterns and improve listener comprehension.

3. Grounds for Unpatentability

Ground 1: Claims 1, 7-8, and 14-15 are obvious over Walker in view of Matsumoto.

• Prior Art Relied Upon: Walker (Application # 2001/0049602) and Matsumoto (Application # 2007/0233492).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Walker discloses a base text-to-speech system that converts text, such as a weather report, into audible speech via a voice application. However, Walker lacks specific implementation details for its TTS engine. Matsumoto discloses a speech synthesizer that improves the naturalness of speech for texts like weather forecasts by comparing sequential sentences (text strings) to identify differing words or phrases. Matsumoto's "collation unit" calculates a "variation degree" for corresponding words; a high degree (different words) triggers an adjustment to prosody (pitch, volume, speed), while a low degree (same words) does not. This combination, Petitioner asserted, teaches identifying a differing first portion and a non-differing second portion between two text strings (claim 1[b]) and assigning contrastive stress to only the differing portion (claim 1[c]).
  • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) implementing Walker’s system would be motivated to find a suitable TTS engine to improve speech naturalness. A POSITA would have readily found Matsumoto, which addresses the same problem (natural-sounding weather reports) and provides the specific mechanism for comparing text strings and applying prosody that is absent in Walker.
  • Expectation of Success: Success would be expected because the combination involves incorporating a known type of TTS engine (Matsumoto) into a known TTS system (Walker) to achieve the predictable result of more natural-sounding synthesized speech.

Ground 2: Claims 2-5, 9-12, and 16-19 are obvious over Walker in view of Matsumoto and Malsheen.

• Prior Art Relied Upon: Walker (Application # 2001/0049602), Matsumoto (Application # 2007/0233492), and Malsheen (Patent 5,634,084).

• Core Argument for this Ground:

  • Prior Art Mapping: This ground builds on the Walker/Matsumoto combination by adding Malsheen to teach the limitation of "normalized orthography" (claim 2). Petitioner contended that weather reports often contain numerals (e.g., "70 degrees," "80 degrees"). While Walker’s system processes such reports, it does not detail how to handle numerals. Malsheen explicitly teaches a text-to-speech converter that improves the handling of text with numbers by using a "Number Expander procedure" to expand numerals into full word strings (e.g., "80" becomes "eighty"). Adding Malsheen’s teaching to the base combination provides the explicit step of converting numerals to words before the Matsumoto collation unit identifies "eighty" as differing from "seventy."
  • Motivation to Combine: A POSITA improving the Walker/Matsumoto system for weather reports would need to address the common presence of numerals. The POSITA would be motivated to incorporate Malsheen's known technique for converting numerals to words to ensure proper pronunciation and enable accurate comparison by Matsumoto’s collation unit.
  • Expectation of Success: The combination would predictably succeed because it applies a standard text pre-processing technique (numeral-to-word expansion from Malsheen) to a known TTS system to improve its functionality for a common data type.

• Additional Grounds: Petitioner asserted several additional obviousness challenges, including combinations incorporating Marple (Application # 2008/0195391) to teach the use of a pre-recorded phoneme database for generating speech, and Bellegarda (Patent 7,313,523) to teach an alternative method of not de-emphasizing unchanged words.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that the Board should not exercise discretionary denial under §314(a) based on the Fintiv factors. Key arguments included that the parallel district court litigation is at a very early stage, with minimal investment and no substantive orders issued. The scheduled trial date is speculative and more than a year away. Furthermore, Petitioner argued that there is incomplete overlap between the proceedings, as this inter partes review (IPR) challenges all 20 claims of the ’671 patent, whereas the district court case involves a smaller subset. Petitioner also argued against denial under §325(d), noting that none of the prior art references presented in the petition were previously considered by the USPTO during prosecution.

5. Relief Requested

• Petitioner requests institution of an IPR and cancellation of claims 1-20 of the ’671 patent as unpatentable under 35 U.S.C. §103.