BPAI Speaks Up on "Transformation" Prong

Ex Parte Hardwick
Decided June 22, 2009

In one of the more confusing sections of the CAFC's In re Bilski opinion, the court gives two examples from a prior case, In re Abele, in attempting to explain what claims should pass the transformation test, and what claims should not, with respect to "electronic signals and electronically-manipulated data."  The Bilski court implied that when a claim included data that "clearly represented physical and tangible objects", then "the transformation of that raw data into a particular visual depiction of a physical object on a display" sufficed to make the claim patent eligible.

On the other hand, the Bilski court noted that Abele found unpatentable a claim that "did not specify any particular type or nature of data" and did not "specify how or from where the data was obtained or what the data represented."

Unfortunately, the CAFC did not give any further examples that would provide guidance between these two extremes.  For instance, it would be nice to have more data points so we might know with more certainty what is necessary versus what is sufficient to render a transformation patentable (or unpatentable). This has been the subject of some discussion in comments on this blog, with various positions being taken.

Today, the Board of Patent Appeals and Interferences decided Ex Parte Hardwick, and in doing so, provided at least some further discussion of the Abele examples raised by the Bilski court.  At issue were claims related to speech processing, rather than image processing.  (The assignee is Westford, Mass.-based Digital Voice Systems, Inc.) After disposing of the Appellant's "machine" prong argument (note how the Board distinguishes computing a "digital filter" from just computing the "coefficients" of a digital filter), the BPAI turned to transformation:

Bilski held that there is no danger of pre-emption of fundamental principles “[s]o long as the claimed process is limited to a practical application of a fundamental principle to transform specific data, and the claim is limited to a visual depiction that represents physical objects or substances.” Bilski, 545 F.3d at 963. Applying this “exception” to the rule requiring transformation of an article to Appellant’s claim 1, we find that claim 1 does specify synthesizing digital speech samples from speech model parameters. Claim 1 does not, however, specify how those parameters were obtained, nor recite transformation of raw data into a visual depiction, or any other analogous depiction (e.g., audio output) that would be cognizable by a human operator. Therefore, the weight of the evidence indicates that Appellant’s claim 1 is substantially more similar to the nonstatutory independent claim in Abele than to the statutory dependent claim. We therefore conclude that the invention of claim 1 is not tied to a particular machine, nor does it transform an article. As a result, we conclude that claim 1 is directed to nonstatutory subject matter.