Prior art for a mobile speech access US20120253817

Question

US20120253817 Mobile speech attendant access, filed Apr 4, 2011

Claim 1 has to do with a mobile device using speech recognition on a remote server to place a phone call and is well known. Claim 2 has to do with "connecting the call to the mobile communication device and the call to the telephone number as an incoming call to the mobile communication device using the telephony server." It's claim 2 that I'm having difficulty with.

1. A method for connecting to a telephone extension listed in a remote telephone directory, comprising:

• recording an audio token on a mobile communication device, wherein the audio token is associated with a telephone number included in the remote telephone directory;
• transmitting the audio token from the mobile communication device to a server over a digital channel;
• matching the audio token received at the server over the digital channel with a telephone number in the remote telephone directory that is associated with the audio token using a speech recognition program; and
• connecting the mobile communication device with a telephone line associated with the telephone number.

2. The method of claim 1, wherein connecting the mobile communication device further comprises:

• calling the mobile communication device and the telephone number using a telephony server; and
• connecting the call to the mobile communication device and the call to the telephone number as an incoming call to the mobile communication device using the telephony server.

Answer 1

Actually, even claim 1 is not as straightforward as it seems at first glance, because it is vague about how the "connecting the mobile communication device with a telephone line associated with the telephone number" step works. It could be interpreted in two ways:

1. If the phone number detected by the server from the audio token is sent back to the mobile device, which then calls the number, then that is along the lines of commonly known prior art, such as voice dialing. (Modern voice dialing actually works by performing speech recognition on the device itself, because the limited vocabulary required allows it. However, shifting the speech recognition to the server is obvious.)

2. But the claim could also be interpreted to mean that the server itself initiates the connection between the caller and the callee, which is what the second claim indicates.

Since the independent claim is vague about it, it should be rejected for 1) indefiniteness, and 2) being non-novel/obvious in light of voice dialing prior art. My guess is, that after the initial rejection, the applicants will modify it to include that interpretation in the first claim. So relevant prior art should probably mention telephony server-initiated calls.

The second claim further says that after you speak the number, you receive a call, which connects you to the number you called. Connecting two numbers so that they both receive an incoming call is a very old capability, and it has been possible since the days when operators manually connected calls. (It's a bit odd to get one like this: you get a call, you pick it up and you hear a ringing dial tone.)

As such, even the second interpretation should be an easy obviousness-type rejection. I have only briefly looked at this area before, but I believe there is tons of relevant prior art.