Method and apparatus for voice message editing

• US 7,706,512 B2
• Filed: 12/29/2005
• Issued: 04/27/2010
• Est. Priority Date: 02/16/2005
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Groeger, Steven, Lister, Jamie, Hayton, Stuart John, Poultney, Timothy David
• Primary Examiner(s): Anwah; Olisa
• Application Number:
  US11/322,032
• Publication Number:
  US 20060182238A1
• Time in Patent Office: 1,580 Days
• Field of Search: 379/88.16, 379/75, 379/88.25, 379/68, 704/252, 704/253, 704/254
• US Class Current: 379/88.16
• CPC Class Codes: G10L 15/00 Speech recognition G10L17/00 takes precedence
• CPC Class Codes: G10L 21/00 Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility G10L19/00 takes precedence

Abstract

A method and apparatus for voice message editing, particularly a method and apparatus for splicing voicemails sections together. A system, method and computer program product are described for forming a voice message on an interactive voice response system (IVR) spoken by a user including recording a first voice message where the user speaks the first voice message but makes an error in a last portion of the first voice message. A second voice message is recorded where the user speaks the last portion again without the error. Splice points are determined in the first and second voice messages. The first and second voice messages are spliced at the splice points such that the spliced first and second voice message is a continuous voice message including the last portion but not including the error.

Claims

1. A method of forming a voice message spoken by a user on an interactive voice response system (IVR) comprising:

prompting the user to record a first voice message and prompting the user to press a first key when the user is finished and press a second key if the user makes an error;

storing the recorded first voice message if the first key is pressed;

prompting the user to record a second voice message if the second key is pressed, wherein the second voice message starts from a point in the first voice message before the error;

when the second voice message is finished, submitting the first and second voice messages for recognition to obtain a first recognition text and a second recognition text;

determining splice points in the first and second voice messages by comparing the first and second recognition texts of the first and second voice messages and determining regions of matching text, wherein the splice points are corresponding points within the regions of the matching text; and

splicing the first and second voice messages at the splice points whereby the spliced first and second voice message is a continuous voice message without the error.

2. The method according to claim 1, wherein each of the first and second recognition texts comprises a phoneme string or human-readable text.

3. A system of forming a voice message on an interactive voice response system (IVR) spoken by a user comprising:

at least one memory for storing recorded voice messages;

at least one processor configured to;

prompt the user to record a first voice message and prompt the user to press a first key when the user is finished and press a second key if the user makes an error;

store the recorded first voice message in the memory if the first key is pressed;

prompt the user to record a second voice message if the second key is pressed, wherein the second voice message starts from a point in the first voice message before the error;

when the second voice message is finished, submit the first and second voice messages to a speech recognition engine for recognition to obtain a first recognition text and a second recognition text;

determine splice points in the first and second voice messages by comparing the first and second recognition texts of the first and second voice messages and determining regions of matching text, wherein the splice points are corresponding points within the regions of the matching text; and

splice the first and second voice messages at the splice points whereby the spliced first and second voice message is a continuous voice message without the error.

4. The system according to claim 3, wherein each of the first and second recognition texts comprises a phoneme string or human-readable text.

5. A computer-readable storage medium, having stored thereon a computer program having a plurality of code sections executable by a computer for causing the computer to perform the steps of:

prompting a user to record a first voice message and prompting the user to press a first key when the user is finished and press a second key if the user makes an error;

storing the recorded first voice message if the first key is pressed;

prompting the user to record a second voice message if the second key is pressed, wherein the second voice message starts from a point in the first voice message before the error;

when the second voice message is finished, submitting the first and second voice messages for recognition to obtain a first recognition text and a second recognition text;

determining splice points in the first and second voice messages by comparing the first and second recognition texts of the first and second voice messages and determining regions of matching text, wherein the splice points are corresponding points within the regions of the matching text; and

splicing the first and second voice messages at the splice points whereby the spliced first and second voice message is a continuous voice message without the error.

6. The computer-readable storage medium according to claim 5, wherein each of the first and second recognition texts comprises a phoneme string or human-readable text.