International Business Machines Corporation

David G Martin

Lawrence Leon Porter

Sanjay Nagchowdhury

Nigel Lewis Jones

Philip Randall Coxhead

A system of controlling performance in an interactive voice response system includes a voice device driver, a voice segment stored in a file in a directory in a standard operating system format; a buffer for storing the voice segment prior to sending to the voice device driver and a plurality of voice channels for output of the voice segment. A sequence of voice blocks is requested to be sent to a buffer, the sequence being one of a plurality of sequences making up a voice segment. The number of voice blocks sent from the file to the buffer is determined. The play period of the sequence and the next underrun time when the sequence will finish playing based on the initiate request time and the play period are calculated. Also calculated is the margin period between the calculated next underrun time and the actual time after a further sequence of voice data blocks is sent to the buffer in response to a device driver request for a further sequence of voice data blocks. A telephony channel is shut down when the margin period exceeds a defined threshold.

Controlling interactive voice response system performance

An interactive voice response application on a computer telephony system includes a method of playing voice prompts from a mixed set of pre-recorded voice prompts and voice prompts synthesised from a text-to-speech process. The method comprises: reserving memory for a synthesised prompt and a pre-recorded prompt associated with a particular prompt identifier; on a play prompt request selecting the pre-recorded prompt if available and outputting through a voice output; otherwise selecting the synthesised prompt and playing the selected voice prompt through the voice output. If neither pre-recorded or synthesised data are available then text associated with the voice prompt is output through a text-to-speech output.

Developing voice response applications from pre-recorded voice and stored text-to-speech prompts

The present invention relates to system for varying the voice menus and segments presented to the user of a voice response system according to the competence of the user. The response time of a user to voice prompts is measured and an average response time is determined. It is assumed that the lower the average response time, the greater the competence of the user. The average response time is used as an index to a table of ranges of response times. Each range has respective voice segments associated therewith. The voice segments comprise oral instructions or queries for the user and vary according to the anticipated competence of the user. If the average response time changes such that the voice segments indexed are different to the current voice segments then a data base containing information relating to user competence is updated to reflect such a change. Accordingly, when the user next interacts with the voice response system a new set of voice segments more appropriate to the user's competence with be played.

Growing community of inventors

Eastleigh, United Kingdom

David G Martin