Samsung Electronics Co Ltd v. Cerence Operating Co

1. Case Identification

• Case #: IPR2024-01267
• Patent #: 9,026,428
• Filed: August 9, 2024
• Petitioner(s): Samsung Electronics Co., Ltd., and Samsung Electronics, America, Inc.
• Patent Owner(s): Cerence Operating Company
• Challenged Claims: 1-3, 7-9, 12-13

2. Patent Overview

• Title: System and Method for Receiving Character Input From a User
• Brief Description: The ’428 patent discloses a method for data input on a mobile device's touchscreen. The system receives handwritten input, determines a best-match recognized sequence, stores it in a memory buffer, and presents it to the user for editing. The core inventive concept asserted involves automatically identifying a "committable sequence" (e.g., a complete word) using linguistic rules and sending that portion to an application while removing it from the buffer, without requiring specific user input like selecting the word or pressing a space key.

3. Grounds for Unpatentability

Ground 1: Obviousness over Arai and Fenwick - Claims 1-3, 7-9, and 12-13 are obvious over Arai in view of Fenwick.

• Prior Art Relied Upon: Arai (JP Publication No. 2006323502A) and Fenwick (Patent 5,161,245).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Arai, a Japanese patent publication, discloses nearly all limitations of the challenged claims. Arai teaches a handwriting recognition system for a mobile device (like a PDA) with a touchscreen. The system receives handwritten characters (e.g., Japanese characters), determines a best match recognition for each, generates a candidate sequence, stores it in a buffer (RAM), displays it to the user for editing, and uses a dictionary for post-processing to correct the sequence. However, Arai requires a user to press an "Enter Key" to send the confirmed text to an application. Petitioner asserted that Fenwick, which describes a pattern recognition system for identifying words from character input, supplies the missing limitation: automatically committing a recognized word and clearing it from the buffer without user intervention. Fenwick teaches analyzing characters in multiple memory buffers and, upon positively identifying a word common to all buffers (based on a dictionary), automatically deleting the word from the buffers and sending it to an output device.
  • Motivation to Combine: Petitioner contended a person of ordinary skill in the art (POSITA) would combine Fenwick with Arai to improve Arai’s functionality. Arai's stated goal was to create a "natural use experience" by reducing extraneous user inputs. Fenwick’s method of automatically committing a confirmed word without requiring an "Enter" key directly addresses and furthers this goal. Combining Fenwick’s automatic commitment process with Arai’s handwriting recognition system was presented as a straightforward application of a known technique to improve a known system.
  • Expectation of Success: A POSITA would have had a reasonable expectation of success in this combination. Integrating an automatic word-commitment module (like Fenwick's) into a character recognition system that uses a temporary buffer (like Arai's) was a predictable design choice that would yield the expected result of a more seamless text entry experience.

Ground 2: Obviousness over Arai, Fenwick, and Baker - Claims 9, 12, and 13 are obvious over Arai in view of Fenwick and Baker.

• Prior Art Relied Upon: Arai (JP Publication No. 2006323502A), Fenwick (Patent 5,161,245), and Baker (EP Publication No. 0830668B1).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground builds upon the combination of Arai and Fenwick from Ground 1 to address the specific limitation in claim 9[e] requiring the matching of a word or phrase in a "database of acceptable words." Petitioner argued that while the Arai/Fenwick combination inherently teaches using a dictionary, the term "database" might require an explicit teaching. Baker provides this teaching. Baker discloses a word recognition system that uses a dictionary to evaluate and replace unrecognized words, and explicitly teaches that this dictionary can be stored in a "reference database."
  • Motivation to Combine: A POSITA implementing the dictionary-based post-processing taught by Arai would need to decide how to store the dictionary data. Petitioner argued that Baker provides a conventional and obvious solution: implementing the dictionary as a database. This would be a simple and well-known design choice for organizing and accessing dictionary data efficiently. The motivation was to apply a known data storage technique (a database) to a known system (the Arai/Fenwick combination) that uses such data.
  • Expectation of Success: Storing a dictionary in a database was a routine and well-understood practice in the art at the time. Therefore, a POSITA would have a high expectation of success in implementing Arai's dictionary within the database structure taught by Baker to achieve a predictable and functional system.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that the Board should not exercise discretionary denial under 35 U.S.C. §314(a) based on the Fintiv factors. Petitioner asserted that while the parallel district court trial is scheduled approximately six months before the projected Final Written Decision (FWD), this factor is not determinative and is outweighed by others. Specifically, Petitioner argued that investment in the litigation is low, and there is an incomplete overlap between the IPR and the court proceeding, as the IPR challenges claims 2-3, 7-9, and 12-13, which are not asserted in the district court case. Petitioner also argued against denial under §325(d), noting that none of the cited prior art was considered during the patent's original prosecution.

5. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1-3, 7-9, and 12-13 of Patent 9,026,428 as unpatentable.