Computer generated prompting
View Patent ↗A method and apparatus for generating appropriate confirmatory prompts in a speech-enabled, interactive computer system. The method can be incorporated in an interactive voice response system that includes receiving an input audio stream over a voice channel from a users, performing keyword recognition on received input audio as subsequent input audio is being received, and prompting the user with an acknowledgement of the keyword or keywords as subsequent input audio is being received. In another aspect of the method, the volume of the speech input can be continuously monitored. In a further aspect of the method, recognition results and associated confidence values can be combined to select different confirmatory prompts, and the volume is tailored to be the same as, louder than or quieter than the volume of the speech input, so that different types of confirmation can be automatically generated to produce a natural speech-enabled interface.
1. A method of providing acknowledgement prompts in an interactive voice response system (IVR) comprising:
receiving an input audio stream over a voice channel from a user;
analyzing elements of the input audio stream by a speech recognition system for identifying a keyword and an associated confidence value;
determining a confidence category based on the identified confidence value;
selecting a suitable acknowledgement prompt associated with the determined confidence category;
while analyzing elements of the input audio stream by the speech recognition system, analyzing elements of the input audio stream by a volume analyzer for determining a current volume;
selecting a suitable output volume for the selected acknowledgement prompt based on the determined confidence category and the current volume; and
outputting to the user the selected acknowledgement prompt at the selected output volume whilst continuing to receive the input audio stream.
2. The method according to claim 1 wherein the voice channel is a full duplex voice channel between a user telephone and an IVR whereby the acknowledgement is played out over the same full duplex voice channel as the input audio stream is being received.
3. The method according to claim 1 wherein for a high confidence value, the form of the confirmation is a “place-marker” prompt.
4. The method according to claim 1 wherein for a medium confidence value, the form of the acknowledgement is an “optional turn-take” prompt plus the identified keyword.
5. The method according to claim 1 wherein for a low confidence value, the form of the acknowledgement is a “definite turn-take” prompt and optionally the identified keyword.
6. The method according to claim 1 further comprising playing out the acknowledgement prompt at the output volume and modifying the output volume prior to being played out.
7. The method according to claim 6 wherein the output volume for the acknowledgement prompt is modified according to the confidence value associated with recognition.
8. The method according to claim 7 wherein the output volume is matched to an input volume or slightly reduced for ambiguous recognition results.
9. The method according to claim 7 wherein the output volume is modified to be louder than an input volume for low confidence recognition results.
10. The method according to claim 7 wherein the output volume is modified to be quieter than an input volume for high confidence recognition results.