`U.S. Patent No. 10,062,377
`
`Oral Argument, March 12, 2024
`
`Apple Inc. v. Zentian Ltd.
`Case No. IPR2023-00035
`
`Petitioner’s Demonstrative Exhibits – Not Evidence
`
`Petitioner’s DX-1
`
`IPR2023-00035
`Apple EX1078 Page 1
`
`
`
`Claim 1 and Proposed Ground 1
`
`- Ground 1: Claim 1
`- Jiang (Ex. 1004) in view of Smyth (Ex. 1005)
`
`Petition (Pet.) (Paper 1), 5, 10-59; ’377 Patent (Ex. 1001), 38:53-39:8
`
`Petitioner’s DX-2
`
`
`
`Jiang Does Not Require Codewords
`
`Jiang teaches
`encoding feature
`vectors from
`spectral
`characteristics
`
`Jiang teaches feature
`vectors further
`encoded into one
`or more codewords
`
`Jiang teaches
`optionally
`outputting FVs or
`codewords for
`distance calculations
`
`Pet., 12, 22, 25, 59; Pet. Reply, 1-2
`
`Jiang (Ex. 1004), 6:62–7:15
`Petitioner’s DX-3
`
`
`
`Zentian’s Arguments Regarding Codewords
`Are Technically Incorrect
`
`Codewords:
`
`• DO contain a plurality of spectral
`characteristics;
`• ARE comprised of quantities derived from a
`digital audio stream;
`• Are NOT a single value;
`
`Distance Calculations from
`Codewords:
`• ARE distance calculations from feature vectors
`
`Pet., 22-27; Pet. Reply, 3, 4-6, 10
`
`Petitioner’s DX-4
`
`
`
`Petition’s Mapping for Calculating Distances
`- Petition details codewords as feature vectors for
`calculating distances
`
`Pet., 22-25; Schmandt Dec. (Ex. 1003), ¶¶ 109, 140-148
`
`Pet. 22-23
`
`Petitioner’s DX-5
`
`
`
`Codewords Represent a Plurality of Cepstral
`Characteristics from a Digital Audio Stream
`
`Speech
`
`Digital Audio
`Stream
`
`Feature
`Vector
`
`Codeword
`
`- “Per Mr. Schmandt, vector quantization is a well-known technique to
`reduce computation requirements and the search space by ‘representing
`the various cepstral features in an efficient manner’”
`- Pet. 22, citing Ex. 1003, ¶¶ 140-141
`- “As noted above, Vector Quantizer (VQ) 110 represents all possible
`cepstral features in an efficient manner. They are then stored in a
`codebook.”
`- Ex. 1036, 10:24-34
`- “Hence the VQ representation is potentially an extremely efficient
`representation of the spectral information in the speech signal.”
`- Ex. 1015, 154
`
`Pet., 12, 22; Pet. Reply, 5, 6; Schmandt Dec. (Ex. 1003), ¶¶ 140-141 (citing Brown (Ex. 1036), 10:24-34,
`3:53-65 and Ravishankar (Ex. 1012), 48-49); Rabiner (Ex. 1015), 154-155, 162-163
`
`Petitioner’s DX-6
`
`
`
`A Codeword Is NOT a Single Value
`
`- Zentian conflates a codeword feature vector with a codeword
`index
`Codeword
`Feature Vector
`- Sur-Reply, 1-3
`
`"Codeword"
`
`Codeword
`Codeword
`Feature Vector
`Index
`- Zentian’s support confirms a codeword is not a single value
`
`Pet., 22-23; Pet. Reply, 3, 4-5; Rabiner (Ex. 1015), 154-155, 162-163; Anderson Depo. Tr. (Ex. 1073), 54:12-
`15
`
`Sur-Reply, 11
`Petitioner’s DX-7
`
`
`
`A Codeword CANNOT be a Single Value
`
`Single Value
`Codeword
`
`compared
`to
`
`N-Dimens.
`Acoustic
`Ac
`stic
`Model
`
`odel MMo lModel
`
`Vector
`
`N-Dimens.
`Codeword
`
`compared
`to
`
`N-Dimens.
`Acoustic
`Model
`Vector
`
`Distance
`Calculation
`
`Distance
`Calculation
`
`Pet., 22-27; Pet. Reply, 5-6, 10; Schmandt Dec. (Ex. 1003), ¶¶ 140-148
`
`Petitioner’s DX-8
`
`
`
`=
`Distance Calculations from FVs
`Distance Calculations from Codewords
`- Distance calculations compare feature vectors,
`representing the cepstral characteristics of a speech
`utterance, to template vectors from an acoustic model
`Distance Calculation
`Acoustic Model vs.
`Acoustic Model vs.
`Input FV
`Codeword
`
`Same Dimensional Size Vectors
`
`A Comparison Between Cepstral
`Characteristics and Acoustic
`States
`A Determination of Similarity
`
`Pet., 22-27; Pet. Reply, 5, 6, 10; Schmandt Dec. (Ex. 1003), ¶¶ 109, 140-148
`
`Petitioner’s DX-9
`
`
`
`Zentian Imposes an Unclaimed Degree of
`Accuracy into the Claims
`- Zentian argues allowing for codewords would broaden
`the scope of the claims “beyond their express language”
`- Sur-Reply, 11
`
`- The feature vector, defined in the ’377 specification, does
`not have any specific measure of accuracy
`
`’377 Patent, 13:19-23
`
`Pet. Reply, 7, 9-12; ’377 Patent (Ex. 1001), 38:53-39:8
`
`Petitioner’s DX-10
`
`
`
`Motivation to Combine Smyth and Jiang
`
`- Ex. 1003, ¶ 183 Includes:
`
`”[A] well-known distance
`calculation would have been
`readily implemented.”
`
`“[P]erforming such a distance
`calculation for determining the most
`likely phoneme would have been
`using a known technique.”
`
`“[T]o improve Jiang’s speech
`recognition system in the same
`way as Smyth’s”
`
`Reasonable
`Expectation of
`Success
`
`“[S]uch a substitution of one
`known element for another
`would have yielded predictable
`results.”
`
`Pet., 40-41; Pet. Reply, 9, 12-13; Schmandt Dec. (Ex. 1003), ¶¶ 170-183
`
`Petitioner’s DX-11
`
`
`
`Increased Computational Load Tradeoff
`
`Pet. Reply, 14
`
`Pet., 39; Pet. Reply, 14; Schmandt Supplemental Declaration (Ex. 1072), ¶ 36
`
`Petitioner’s DX-12
`
`
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
