Opinion
Patent Appeal No. 9249.
June 27, 1974.
Nathan Cass, Westlake Village, Cal., atty. of record, for appellant.
Joseph F. Nakamura, Washington, D.C., for Commissioner of Patents. Jere W. Sears, Washington, D.C., of counsel.
Appeal from the Board of Appeals.
Before MARKEY, Chief Judge, and RICH, BALDWIN, LANE and MILLER, Judges.
This appeal is from the decision of the Board of Appeals affirming the rejection, for insufficiency of disclosure under 35 U.S.C. § 112, first paragraph, of all the claims (claims 1, 2, 4, 5, 7 and 8) remaining in appellant's application serial No. 768,739, filed October 18, 1968, for "Computer-Controlled Vocal Response." We affirm.
A preliminary matter, a motion by the Commissioner of Patents to strike U.S. Patent No. 3,479,648 to Sleven et al. from the record, is also before us. The patent was added to the transcript of record by motion of appellant (over objection by the Commissioner which was denied without prejudice to renewal of such objection) for the purpose of showing that appellant's specification was enabling. However, the patent issued November 18, 1969, after appellant's filing date, and hence cannot be relied on as evidence for that purpose. In re Glass, 492 F.2d 1228 (CCPA 1974). Therefore, the motion to strike is granted.
The Invention
The invention relates to providing vocal responses from a vocabulary of spoken words stored in analog form under the control of a computer.
The sole drawing is set forth below: [numeral added in bracket]
An explanation of the invention suitable for present purposes is given in appellant's brief as follows:
Briefly, in accordance with the invention, a large vocabulary of spoken words are recorded in analog form at a relatively low speed on an appropriate record medium, such as the drum memory 10 illustrated in the drawing. In order to obtain compact storage of the words constituting the vocabulary, a large plurality of words are recorded on each of the 64 tracks or channels which may typically be provided on the drum. In composing a desired message for vocalization, selected words constituting the desired message are played back from the drum at a relatively higher speed than was used for recording. During this high speed playback, a conventional general purpose digital computer 11 cooperates with a head selection network 12, a sector selection gate 16 and a timing and control section 13 to provide for selection and read out from the drum of the particular analog spoken words required for the desired message. The selected spoken words thus read out from the drum are fed to a high speed A/D converter 17 for conversion to equivalent digital form. The resulting digitized spoken words are suitably assembled by group assembly 18 and then transferred via a date register 19 for storage in the high speed buffer core memory of the computer 11.
Once the digitized selected spoken words constituting the desired message have been stored in the computer memory 11, they are then available for transfer to the data register 25 of an appropriate output channel 21 or 22 via input gates 26 under the control of an output control section 24 provided in each channel. A desired controllable delay between words read out from the computer memory is provided using a delay counter 23 in the channel 21 which is preset by the computer 11 and is counted down to provide the desired delay between words. The digitized spoken words of the desired message thus read out into the channel data register 25 from the computer memory are then applied via a bank of disassembly gates 27 to a D/A converter 28 for reconverting the digitized spoken words into analog form. The reconverted spoken words are then fed via a filter 30 to an audio device 29 which provides for vocalization of the desired message.
Thus, the present invention not only makes possible more compact analog storage of the spoken words constituting the system vocabulary, but also offers the very significant additional feature of making it possible to take advantage of digitizing techniques in a spoken word vocalizing system without suffering the disadvantages that would occur if the words were placed on the drum in equivalent or abbreviated digital form in the first instance. In particular, it is to be noted that digitizing is not required in appellant's system until the selected analog-recorded spoken words constituting the desired message are read from the drum, which can be done at relatively high speed as compared to the recording speed, since the message is not vocalized during such playback. Since only relatively few spoken words on the drums are selected for playback in order to compose a desired message, the digital computer core memory need provide storage only for these relatively small number of selected digitized spoken words, rather than for the entire vocabulary.
The prior art systems for producing a selected vocal message with which the last paragraph quoted above is concerned are described in appellant's brief to be of three types as follows:
* * * direct (i.e., analog) recording of spoken words in analog form, (for example, as occurs in a conventional commercial tape recorder) on selectable tracks of a drum or like storage medium, with each track preferably storing one word. A desired vocal response is provided from these recorded analog signals by sequentially playing back, in an appropriate sequence, selected ones of the recorded spoken words at the same speed as used during recording.
* * * * * *
* * * the storage of the vocabulary of spoken words in digital form rather than in analog form, so as to thereby permit taking advantage of digitizing techniques in composing the desired message to be vocalized. In such a system, the vocabulary of spoken words are stored in equivalent digital form in a suitable memory and appropriately selected for composition of the desired message to be spoken.
* * * * * *
* * * systems [which] attempt to obviate the need for a large memory for digital storage of the words of the vocabulary by storing only abbreviated digital representations of each spoken word instead of the equivalent digital representation thereof. * * *
Although the application includes two method claims in addition to system claims, claim 4 is considered representative:
4. A system for providing controlled vocal responses comprising:
recording means including a memory device having a multi-track cyclically moving record-medium for storing a vocabulary of spoken words recorded in analog form in the tracks of the record-medium at a relatively low speed and with a large plurality of spoken words being recorded in each track;
first means for selectively playing back words from said memory device at a very much higher speed than said spoken words were recorded, each word being played back in sequence;
second means connected to said first means for converting each selected word from analog to an equivalent digital form as it is played gack in sequence;
third means connected to said second means for storing each selected word in digital form as it is converted in sequence;
fourth means connected to said third means for transmitting each selected word stored in digital form at a low speed commensurate with the speed at which the selected word was recorded in said memory device;
fifth means connected to said fourth means for converting each selected word in sequence from digital to analog form as it is transmitted by said fourth means; and
an audio device connected to said fifth means for producing an audible sound of each selected word in sequence from its analog-signal form.
The Rejection
The rejection sustained by the board, as previously indicated, is under 35 U.S.C. § 112, first paragraph. The board concluded that the examiner regarded the disclosure as insufficient to enable one to make or use the invention,
with particular reference to appellant's "DIGITAL-COMPUTER" 11, "TIMING AND CONTROL" 13, "A/D CONVERTER" 17, "DELAY COUNTER" 23, and "OUTPUT CONTROL" 24 to which obviously should have been included the "D/A CONVERTER" 28.
It was the board's view that appellant, by identifying each of the above enumerated devices only by generic name and the overall ultimate function that each is intended to perform, had represented that from this alone "a person of ordinary skill would be enabled to make and use the apparatus with only a reasonable amount of experimentation." The board further noted that the examiner had challenged the disclosures of these enumerated devices, "which challenges we find reasonable on their face." It considered the appellant to have been "put to his proof" that his application was sufficient to put a person of ordinary skill in the art in possession of the invention.
Commenting that appellant had sought to meet this burden by reference to various features in five patents listed — Williams et al., Avakian et al., Clapper, Koeijmans and Brown — the board stated:
U.S. Patent 3,037,076, granted May 29, 1962.
U.S. Patent 3,133,268, granted May 12, 1964.
U.S. Patent 3,183,303, granted May 11, 1965.
U.S. Patent 3,344,407, granted September 26, 1967.
U.S. Patent 3,366,924, granted January 30, 1968.
We find this evidence wholly unconvincing. First, even if each of the enumerated devices were old, per se, this would not make it self-evident how each would be interconnected to function in the complex manner described in this application. In this complexity, timing between the various units is of the essence and without a timing chart relating the timed sequences for each unit an unreasonable amount of work would be required to come up with the detailed relationships appellant says that he has solved. Secondly, we find that many of the units referred to in the prior art patents have no better detail of disclosure than his own, being labeled boxes with only attributed functions. Furthermore, some of the items in the referred patents clearly are not germane to appellant's devices, as for example, the seismic analog-to-digital converter of the reference do [sic] not appear to be the same as the voice-to-digital converter of the instant application. There is no evidence in the record that these converters are interchangeable.
We note the court in the decision on In re Forman, [59 CCPA 1178, 463 F.2d 1125,] 175 USPQ 12 [1972], stated:
However, it is not made clear [Affidavits of record] that a person skilled in the art would know which, or what parts, of these circuits could be used to construct appellant's device or how they could be interconnected to act in coordination to produce the required results. In fact, it does not appear that the descriptions in the extracts would themselves meet the disclosure requirements of § 112. [Bracketed matter found in board's opinion.]
It is not enough that a person skilled in the art, by carrying on investigations along the line indicated in the instant application, and by a great amount of work eventually might find out how to make and use the instant invention. The statute requires the application itself to inform, not to direct others to find out for themselves, In re Gardner et al., 57 CCPA 1207, 427 F.2d 786, 166 USPQ 138 [1970].
Referring in more detail to the A/D and D/A converters, the board held:
* * * [I]t is self-evident that a disclosure of the conversion apparatus between voice and digital data becomes of the essence to a determination of both the adequacy of the instant disclosure to comply with the statute as well as to any determination of the patentability of the claims over prior art methods and apparatus.
The board reversed a rejection of the claims on the prior art under 35 U.S.C. § 103.
On the basis alone of the complete absence of any disclosure in the application of the apparatus required to supply the essential functions attributed to the converters 17 and 28, and the non-pertinency of the seismic data converters advanced by appellant, the rejection on inadequate disclosure clearly would be supportable.
OPINION
We are in full agreement with the board that the claims were properly rejected for insufficiency of disclosure under the first paragraph of section 112. The sufficiency of the disclosure was continually under challenge from the first action of the examiner. That challenge was plainly adequate to put appellant to his proofs that the disclosure would be enabling to a person skilled in the art. We consider the response of appellant clearly inadequate for the reasons specified by the board.
The arguments made by appellant before us are not persuasive. He argues that "all of the functions required by blocks 11, 13, 17, 23, 24 and 28 are readily providable in the art" and goes on to state that all of the various steps of his process, expressed in broad terms, are known. However, that does not answer the question whether complex elements known to perform broadly recited functions in different systems would be adaptable for use in appellant's particular system with only a reasonable amount of experimentation. Arguments made regarding the A/D and D/A converters of Koeijmans and Brown fail to answer factually the board's justifiable doubts over the applicability of seismic data conversion to the voice sound conversion here involved. As to the matter of the interconnections required between the blocks representing the elements and the timing there between, appellant merely argues that the board is wrong. However, argument of counsel cannot take the place of evidence lacking in the record. In re Cole, 326 F.2d 769, 51 CCPA 919 (1964); In re Schulze, 346 F.2d 600, 52 CCPA 1422 (1965).
The board in this case has interpreted and applied the law almost precisely as did this court in the later case of In re Brandstadter, 484 F.2d 1395 (CCPA 1973), where it was held that an application for a complex message retrieving system did not satisfy the enablement requirement of 35 U.S.C. § 112. As in Brandstadter, we are not convinced that the examiner's challenge to enablement was unreasonable or has been met. We agree with the board that an unreasonable amount of work would be required to arrive at the detailed relationships appellant says that he has solved. See also In re Ghiron, 442 F.2d 985, 58 CCPA 1207 (1971).
In re Marzocchi, 439 F.2d 220, 58 CCPA 1069 (1971), relied on by appellant, involved a different type of situation. Claims 6 and 12, which recited the generic term "polyethyleneamine" as the amine adhesion enhancer, had been rejected under § 112, first paragraph, as being based on a disclosure which was not enabling. This court regarded the rejection as going to the "breadth" of the claim — whether all polyethylene amines possessed the necessary capability of enhancing adhesion. The rejection was reversed because the Patent Office had not presented sufficient reasons to doubt the objective truth of the broad disclosure contained in appellant's specification. In the instant case, there is no challenge to the breadth of the claims nor to the objective truth of the disclosure contained in appellant's specification. What is challenged is the sufficiency of appellant's specification to enable a person skilled in the art to make and use the invention without undue experimentation. We agree with the board that there are adequate grounds for doubting the sufficiency of the present specification.
The decision of the board is affirmed.
Affirmed.