T 2493/11 (Manipulation of audio signals/DMT LICENSING) of 20.1.2017

Summary of Facts and Submissions

I. The applicant (appellant) appealed against the decision of the Examining Division refusing European patent application No. 98907658.3, filed as international application PCT/US98/03840 and published as WO 98/43405.

II. The Examining Division decided that the subject-matter of the independent claims of the sole request on file lacked inventive step. No document was cited as evidence for the relevant state of the art.

III. With the statement of grounds of appeal, the appellant submitted a main request with claims closely similar to those underlying the contested decision and four auxiliary requests. While the appellant argued that the contested decision was wrong, it explicitly accepted that the state of the art relied upon by the Examining Division was known.

IV. In a communication under Article 15(1) RPBA following a summons to oral proceedings, the Board expressed inter alia the provisional opinion that the subject-matter of claim 1 of the main request contained added subject-matter violating Article 123(2) EPC.

In its communication, the Board cited the following documents:

D4: US 5,600,316, published 4 February 1997

D5: US 4,280,192, published 21 July 1981

D6: US 5,590,108, published 31 December 1996

D7: US 5,388,181, published 7 February 1995

and raised objections under Article 56 EPC based on the prior art described in the application itself in combination with either the common general knowledge or any of the documents D4 or D5.

V. With a letter dated 8 November 2016, the appellant maintained all requests unamended and informed the Board that it would not attend the oral proceedings. No arguments addressing the objections raised by the Board were submitted.

VI. Oral proceedings were held as scheduled in absence of the appellant. At the end of the oral proceedings, the chairman pronounced the Board's decision.

VII. Claim 1 of the main request reads as follows:

"1. A method for manipulating digital audio signals characterised by the steps of:

filling a frequency/amplitude database (90) with sound entries of frequency and amplitude information for corresponding time intervals of the digital audio signals (60);

identifying patterns of consecutive sound entries over the time intervals for a specific amplitude of a discrete frequency;

filling a static audio file (110) with a starting point of a specific amplitude of a discrete frequency and its related end point with respect to time over the time intervals, such that the static audio file provides a digital file format which provides a plurality of discrete frequency/amplitude information and the respective starting points and end points;

providing a static audio player (120) with a starting point of a specific amplitude of a discrete frequency with respect to time from the static audio file (110);

playing the specific amplitude of the discrete frequency with the static audio player (120) when the starting point occurs;

providing the static audio player (120) with an ending point of the specific amplitude of the discrete frequency with respect to time from the static audio file (110); and

stopping the playing of the static audio player (120) of the specific amplitude of the discrete frequency when the ending point occurs."

VIII. Claim 1 of the first auxiliary request differs from claim 1 of the main request in that it adds the following feature to the step of filling a frequency/amplitude database:

"the digital audio signals (60) comprising a string of consecutive time intervals which repeat one or more identical frequencies and their related amplitudes".

IX. Claim 1 of the second auxiliary request differs from claim 1 of the first auxiliary request in that it adds the following feature at the end of the step of filling a static audio file:

"without information related to repetition between each starting point and its end point".

X. Claim 1 of the third auxiliary request differs from claim 1 of the second auxiliary request in that it adds the following feature at the end of the step of playing the specific amplitude:

"and continuing to play the specific amplitude of the discrete frequency during subsequent time intervals until a time interval corresponding to its ending point occurs".

XI. Claim 1 of the fourth auxiliary request differs from claim 1 of the third auxiliary request in that it adds the following text to the step of filling a frequency/amplitude database after "comprising":

"frequency/amplitude information by time interval and".

The wording of the further claims of the requests is not relevant to the present decision.

XII. The appellant's arguments relevant to the decision are discussed in detail below.

Reasons for the Decision

1. The appeal complies with the provisions referred to in Rule 101 EPC and is therefore admissible.

2. The invention

2.1 The application relates to a method and system to compress and/or convert audio signals into a specific file format called "static file format" which can then be played back by a static audio player.

According to the background of the invention, known computer file formats for digital audio, which are formatted in a dynamic manner permitting easy and routine editing, result in very large computer file sizes (for example, hundreds of millions of bytes in size for a 40 minute digital audio file of 44.1 kHz sound quality). The very large file sizes of these "Dynamic Audio Files" are problematic for data transmission and storage.

2.2 The application provides an example of a known audio file format on page 2, lines 1 to 18: Each second of a CD quality Dynamic Audio File is divided into 44,100 discrete time intervals. Each of these time intervals can simultaneously contain multiple frequencies (i.e. pitch) of sound at multiple amplitudes (i.e. volume). The Dynamic Audio File instructs an audio playing device to play discrete frequencies/amplitudes at a rate of 44,100 times per second for CD quality sound. In a Dynamic Audio File, even if a string of consecutive time intervals contains identical frequencies and their related amplitudes, such an occurrence is irrelevant since the Digital Audio File format was designed, in part, to enable specific editing and/or dynamic manipulation of each individual time interval. The Dynamic Audio File fails to take advantage of redundancies within a string of consecutive time intervals which happen to repeat one or more identical frequencies and their related amplitudes.

2.3 The invention proposes to encode audio files in a static format which takes advantage of consecutive redundancies within Dynamic Audio Files with respect to time. A Static Audio File according to the invention is encoded in a format which records a plurality of discrete frequency/amplitude (sound) information to be played on an audio output device, and the respective starting and end points each such frequency/amplitude is to be played, with respect to time. This allows to reduce consecutive redundancies within Dynamic Audio Files due to repetition of frequency/amplitude information. Moreover, the invention proposes a static audio player for playing back digital audio signals recorded in the static audio file on an audio output device (description, page 5, line 1 to page 6, line 10).

Main request

3. Added subject-matter - Article 123(2) EPC

3.1 According to the statement of grounds of appeal claim 1 is based on original claim 3 with several amendments. One of these amendments recites that the step of filling the static audio file is "such that the static audio file provides a digital file format which provides a plurality of discrete frequency/amplitude information and the respective starting points and ending points". In the statement of grounds of appeal the appellant submitted: "Support for this amendment is found at lines 14 to 18 of original page 23." The cited passage reads as follows:

"The Static Audio file 170 is encoded in a digital file format which records a plurality of discrete frequency/amplitude information and the respective starting points and ending points, with respect to time and can be produced by the Static Audio Player 120. The static Audio [...]".

3.2 The Board is not convinced that there is a basis for the phrase "the static audio file provides a digital file format" (emphasis added). The description states that the static audio file is encoded in a digital file format (see description, page 23, line 14). The expression "the static audio file provides a digital file format" can be interpreted in the sense that the file contains information about the digital file format and/or in the sense that the file itself is provided in a specific digital format. The passage cited by the appellant does not support the presence of information about the file format in the static audio file. Nor is the Board aware of any other passage in the original application supporting this amendment.

3.3 As the amendment is not supported by the original application, claim 1 does not comply with the requirements of Article 123(2) EPC.

4. Inventive step - Articles 52(1) and 56 EPC

4.1 Interpretation of claim 1

Claim 1 does not specify whether or how the digital audio data has been obtained from an analog audio signal. In the Board's opinion, the method is applicable to any kind of digital audio data containing frequency and amplitude information.

Claim 1 does not specify in detail the frequency and amplitude information in the sound entries that are filled into the database or the structure of the database. In particular, it is not limited to sound entries containing a specific amplitude of a discrete frequency, or to a specific representation in a three dimensional matrix (description, page 28, lines 16 to 21).

Claim 1 does not specify how the start and end points are obtained or which "patterns of consecutive sound entries over the time intervals" are identified. In the light of the description (for example, page 29, line 22, to page 31, line 10), the Board interprets the patterns as repetitions of a specific amplitude of a discrete frequency in consecutive time intervals.

Claim 1 refers to a "static audio player", which is not a known type of player and thus has no well-known technical meaning. According to the description, page 19, lines 1 to 6, this player is a computer software program running on a conventional computer system used for playing back the static audio file through the sound card of the computer system. The Board interprets the expression "static audio player" in this sense.

The Board interprets the above identified feature infringing Article 123(2) EPC in the light of the description, page 23, line 14, in that the static audio file is encoded in a digital file format.

4.2 The prior art

The Board notes that the application discloses on page 15, lines 17 to 23, that the Dynamic Audio File is already commercially available. Hence, it is part of the prior art. According to page 25, line 6 to page 28, line 8 such a Dynamic Audio File represents sound entries for time intervals by means of specific amplitudes of discrete frequencies. With respect to the file format of these Dynamic Audio Files the application sets out the following on page 25, lines 9 to 21: "...and assuming that a Dynamic Audio File 60 is composed of a plurality of discrete sounds identified by their frequencies and their related amplitudes are mathematically expressed as time interval (I), frequency (F), and amplitude (A), where Iw identifies a discrete time interval within a range of intervals identified by subscript 'w', and bounded by the first time interval and the last time interval of the audio recording; and Fx identifies a discrete frequency within a range of frequencies identified by subscript 'x'; and Ay identifies a specific amplitude, associated with said frequency Fx, within a range of amplitudes identified by subscript 'y'; ... ".

The Board views this known "Dynamic Audio File" as closest prior art which discloses at least the following feature of claim 1:

"a frequency/amplitude database (90) with sound entries of frequency and amplitude information for corresponding time intervals of the digital audio signals (60)".

The Board is of the opinion that the term "database" in a broad sense means any collection of data and that the Dynamic Audio File implements such a collection of data.

The Board refers also to documents D6 (see description, column 7, line 43 to column 8, line 37) and D7 (see description, column 7, line 48 to column 8, line 7), which disclose that digital representations of audio signals in the frequency domain are well known. Consequently, the Board is of the opinion that the claimed frequency/amplitude database representation of audio files was known.

4.3 Technical problem

As set out by the applicant in the description on page 1, last paragraph, dynamic audio files have unfortunately very large file sizes. Hence, the technical problem that a skilled person needs to solve is how to provide compression of such files.

4.4 The solution

According to the method of claim 1, the solution consists essentially in the steps of:

"identifying patterns of consecutive sound entries over the time intervals for a specific amplitude of a discrete frequency;

filling a static audio file (110) with a starting point of a specific amplitude of a discrete frequency and its related end point with respect to time over the time intervals, such that the static audio file provides a digital file format which provides a plurality of discrete frequency/amplitude information and the respective starting points and end points;"

The further steps in the claim relate to the static audio player which is merely using the static file format for reproduction. For a skilled person it is obvious how to use the content of the static audio file in order to reproduce the sound on the static audio player: for a specific amplitude of a discrete frequency the starting and end points available in the static audio file provide the information during which time intervals the specific amplitude of the discrete frequency needs to be played. It follows that the playing back with the static audio player is straightforward.

4.5 Hence, the essential question to be considered is whether the identification of the patterns and the filling of the static audio file, which represent the essential aspects of the compression/conversion activity, involve an inventive step.

4.6 In the Board's opinion, it was well known at the priority date of the present application to compress files by eliminating redundancy. Repetitive values are a generally known form of redundancy. Hence a skilled person, who is familiar with compression techniques, would consider identifying patterns of repeating values and representing time series data (in the present case sound entries representing time intervals of the digital audio signals) of repeating values by the starting and end points of the repeated values in the time series.

4.7 In the opinion of the Board, a skilled person would also be directly led to the claimed solution by each of the documents D4 and D5. These documents belong to the prior art of compression methods for digital representations of audio signals so that a skilled person would consult them in search of a solution to the problem posed.

Document D4 discloses the compression of repeated values by identifying patterns and the use of the identified patterns for encoding the occurrence of repetition, its characteristics and its duration (see D4, abstract; claims 1 to 6; column 3, line 4 to column 4, line 6; column 4, line 28 to column 5, line 44; column 6, lines 15 to 20; column 12, lines 33-49; Figures 1, 8a to 8b).

Document D5 is similar to document D4 and discloses compression for coded audio files by replacement of a pause with start and end times and length of the pause (see D5, abstract, claim 1), and by eliminating repetition of digital audio signal values (see D5, abstract, claims 3 to 6 and 10).

Hence, in the Board's opinion, the claimed steps of identifying patterns and creating static audio files, wherein the identified patterns are used to compress repeated values by representing only their start and end points, are obvious in view of the common general knowledge or in view of the teachings of any of documents D4 or D5.

4.8 Before receipt of the Board's communication pursuant to Article 15(1) RPBA, the appellant submitted various arguments in favour of inventive step. The appellant developed its arguments starting from MPEG audio compression as prior art, which is a standard for lossy compression of audio. It argued that the invention had the technical effect of an increased data efficiency with reduced processing complexity. In particular, the invention recognized the very significant potential for lossless compression by taking advantage of redundancies due to repetition in real-life audio.

However, these arguments are necessarily not pertinent as the appellant could not take into account the Board's newly cited prior art documents D4 and D5 and the new line of arguments based on common general knowledge and the prior art described in the application itself. Consequently, the appellant's arguments based on a comparison with MPEG cannot convince the Board. Moreover, as pointed out above, it was common general knowledge to use compact coding of repetitive audio data in order to achieve data compression.

4.9 In view of the above, the Board concludes that the subject-matter of the independent claims does not involve an inventive step.

First auxiliary request

5. Article 123(2) EPC

5.1 Claim 1 of the first auxiliary request has been amended by adding the phrase "the digital audio signals (69) comprising a string of consecutive time intervals which repeat one or more identical frequencies and their related amplitudes" to claim 1 of the main request.

5.2 The appellant cited page 2, lines 15 to 18 as basis for the amendments. The Board accepts that the amendment conforms to the requirements of Article 123(2) EPC.

6. Article 56 EPC

6.1 In the statement of grounds of appeal, the appellant submitted that the amendment clarified that the invention applied to digital audio signals having redundancies of the type that the invention was configured to remove.

The Board agrees that the amendment improves the clarity of the claim. However, in the Board's opinion, the amendment does not change the assessment of inventive step. The amended feature does not modify the claimed compression scheme, but clarifies merely that redundancies are actually present in the digital audio data. Hence, claim 1 of the first auxiliary request is obvious in view of the general knowledge of a skilled person (or any of the documents D4 or D5).

Further Auxiliary requests

7. Added subject-matter - Article 123(2) EPC

7.1 Claim 1 of each of the second, third and fourth auxiliary requests has been amended by adding the phrase "without information related to repetition between each starting point and its end point" to the step of filling a static audio file.

7.2 The appellant cited page 30, lines 19 to 22 as basis for the amendments.

7.3 The cited passage reads: "Next, the Dynamic to Static Audio Truncator 100 erases all occurrences of sound information related to said repetition strings of frequencies/amplitudes FxAy between the 'on' code and the 'off' code."

7.4 In the Board's view, the cited passage discloses that specific information is erased, whereas the newly added feature states that the file contains no information related to repetition. In fact, the static audio format contains information related to repetition in the form of the "on" and "off" codes which specify that some information is repeated between the starting and end points (see for example the description, page 30, lines 22 to 25, and page 32, line 16 to page 33, line 4).

7.5 Consequently, the amendment is not supported by the application as originally filed and none of the second, third and fourth auxiliary requests meets the requirements of Article 123(2) EPC.

Conclusion

8. As none of the appellant's requests can form the basis for the grant of a patent, the appeal has to be dismissed.

Order

For these reasons it is decided that:

The appeal is dismissed.