Speech recognition optimization tool
AssigneeInternational Business Machines Corporation
InventorsFrancis Fado, Peter J. Guasti
Life of this patent
Estimating the term…
Abstract
A method of optimizing audio input for speech recognition applications can include identifying a source waveform and at least one optimization parameter, wherein the optimization parameter is configured to adjust audio input to a speech recognition application. The source waveform can be modified according to the optimization parameter resulting in a modified waveform. At least one optimization parameter can be synchronized with the source waveform. At least two time dependant graphs can be displayed, where the time dependant graphs can include the source waveform, the modified waveform, and/or a graph for the optimization parameter plotted against time.
Claims24
A method of optimizing audio input for speech recognition applications comprising the steps of:
- identifying a source waveform;
- identifying at least one optimization parameter, wherein said optimization parameter is configured to adjust audio input to a speech recognition application;
- modifying said source waveform according to said optimization parameter resulting in a modified waveform;
- synchronizing said at least one optimization parameter with said source waveform;
- and,
- displaying at least two time dependant graphs selected from the group consisting of said source waveform, said modified waveform, and a graph for said at least one optimization parameter.
A system for optimizing input for speech recognition applications comprising:
- an audio input device containing at least one adjustable input parameter, wherein said audio input device is configured to receive audio input for a speech recognition application;
- a signal editor configured to modify a source waveform by said adjustable input parameters producing a modified waveform;
- a means for converting each of said adjustable input parameters into a parameter waveform, wherein said parameter waveform is a graphical representation of the value of a selected one of said input parameters plotted against time and synchronized with said source waveform and said modified waveform;
- and,
- a means for simultaneously displaying said parameter waveform, said source waveform, and said modified waveform.
A machine-readable storage having stored thereon, a computer program having a plurality of code sections, said code sections executable by a machine for causing the machine to perform the steps of:
- identifying a source waveform;
- identifying at least one optimization parameter, wherein said optimization parameter is configured to adjust audio input to a speech recognition application;
- modifying said source waveform according to said optimization parameter resulting in a modified waveform;
- synchronizing said at least one optimization parameter with said source waveform;
- and,
- displaying at least two time dependant graphs selected from the group consisting of said source waveform, said modified waveform, and a graph for said at least one optimization parameter.
Dependencies are inferred from the claim text.