Insights High Court finds Sky Q system within TV network patent for speech recognition system, but patent invalid



Promptu Systems Corporation issued proceedings against Sky UK Ltd for infringement of its European Patent (UK) entitled, “System and Method of Voice Recognition Near a Wireline Node of Network Supporting Cable Television and/or Video Delivery” (the Patent). Promptu said that Sky’s “Sky Q” subscription TV service infringed the Patent.

Sky denied infringement and counterclaimed that the Patent was invalid for obviousness over a US patent, “Houser”, and should be revoked.

Promptu applied to amend Claim 13 of the Patent, leaving it as the only claim in contention as Promptu did not seek to defend the other claims.

The Patent taught that each user has a remote control and a set-top box. All the set-top boxes in the system communicate over a network to a central server. Each user can give speech commands using microphones in their remote control using  a “push-to-talk” button.  The Patent teaches that “The push-to-talk button may begin the process of speech recognition by informing the system that the subscriber is about to speak and also to provide immediate address information. Address information identifies the user site at which the speaking occurs”. At the central server, speech processing is done centrally for all users. In response to the speech commands, the server sends back customised entertainment and/or information. One source of information used is a “content engine” in the network, which is essentially a database of what content is available along with various meta-data relating to the content (e.g. genre, rating).

The Sky Q system includes several connected systems, which are all connected via the internet, including servers, an Automatic Speech Recognition (ASR) function, and a content engine containing searchable metadata for the content.

When a Sky Q user presses the voice button on their remote control, a persistent connection between the user and the server, individual to the user, is opened. This involves sending, among other things, the user’s IP address to the server. When multiple users activate voice command at the same time, the server receives data from all of them and separates it out into each user’s speech using their IP addresses. The server sends the response back to the user, once it has received what it needs from the ASR and content engine.

The main issues in the case were:

  1. two issues of claim construction;
  2. two infringement issues, namely whether the operation of Sky Q fell within the method claim, and whether there was any infringement at all given some elements of the Sky Q operation took place outside the UK;
  3. obviousness over Houser, the remaining prior art citation; and
  4. two points on added matter.


For context and ease of reference, it is useful to repeat here the individual features of Claim 13 as done by Meade J in his judgment:

Patent construction

Mr Justice Meade found against Sky on construction, holding that Claim 13 as proposed to be amended was not limited to closed TV cable networks and did not exclude networks that used the internet. Although the system described was a cable system of a specific kind, the patent covered other network types or physical set-ups, including the internet, as long as the network was capable of supporting the types of content described in feature 1C (live television and on-demand video content).

Meade J also had to consider whether a particular feature, feature 13, covered sending data as part of making an initial connection to the server or was limited to sending data after a connection had been set up, as argued by Sky. He concluded that there was no basis for Sky’s argument that there had to be a pre-existing connection.


Given the findings on construction, Meade J found that Sky Q fell within the scope of the Claim.

On the question of territoriality, not all parts of the Sky Q system were in the UK as the content engine was hosted in Ireland, and in the case of the ASR, Sky did not know for any given instance whether it was performed in the UK or not. Therefore, some of the steps of Claim 13 were performed in the UK and others were not.

Meade J said that all features of the Claim, except the ASR and content engine access, took place in the UK. The method was, overall, a method that took place at the “server end”, remote from the user. User input triggered part of the method, but it was not the user who put the method into effect.

Content engine access and ASR were subordinate to the central part of the processing at the server end and, in the Sky Q system, were sub-contracted. It was therefore a matter of indifference where they took place, as evidenced by Sky not knowing where they took place. Overall, therefore, Meade J found that the method was performed, in substance, by Sky in the UK.

The Sky Q system therefore infringed the Patent.


Meade J rejected Sky’s added matter arguments but found that the Patent was invalid over Houser.

Houser was entitled: “Information System Having a Speech Interface”. Its area of interest was more specifically subscription TV systems, video on demand, electronic guides and schedules and, in those contexts, speech command. It contained several worked embodiments that had a variety of set-ups, and some had microphones in a remote control.

The two main features of Houser in terms of this case were the references to: (i) a “threshold element”, which it taught could be provided as a power-saving feature to sense when the sound level exceeded a certain level; and (ii) an optional feature of a press to speak button on the remote control.

Sky argued that it was obvious to: (i) choose to use the press to speak button; and (ii) send an alert to the server when the press to speak button was pressed, to cut down on latency.

Promptu argued that: (i) the threshold feature in Houser would be understood to be important, that its inclusion would be seen as inconsistent with the press to speak button, and that the former would be preferred over the latter; and (ii) in relation to an alert being sent on pressing the speak button, Houser did not teach it, nor did it show any sign of even considering it; further, it was not supported by the common general knowledge, nor was it obvious, and there were numerous other ways to deal with latency.

Meade J rejected Promptu’s first argument, as: (i) it was not grounded in the expert evidence; (ii) both features were presented in Houser as options only; (iii) the fact that the threshold was mentioned first in Houser was of no real significance to the skilled reader; and (iv) the features were not inconsistent: using the press to speak button would prevent activity until the button was pressed; once the button was pressed, it would be possible for speech to have an effect, but the threshold would still be useful to prevent problems being caused by background noise below the threshold.

As for the second argument, Meade J agreed with Sky that latency was something the skilled addressee would have in mind. It was not just obvious but inevitable that a careful but uninventive designer of a system of the kind taught in Houser would have to assess where latency would occur and what would contribute to it. Two sources of latency would be: (i) establishing a connection; and (ii) preparing the resources necessary to process speech once it arrived.


Meade J found that proposed amended Claim 13 was obvious over Houser and, because no other Claim was defended, the Patent was invalid and should be revoked. (Promptu Systems Corporation v Sky UK Ltd [2021] EWHC 2021 (Pat) (19 July 2021) — to read the judgment in full, click here).