Internet telephone voice mail management

• US 20070133757A1
• Filed: 12/12/2005
• Published: 06/14/2007
• Est. Priority Date: 12/12/2005
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Kirkland, Dustin, Girouard, Janice, Simon, Kimberly, Ratliff, Emily
• Granted Patent: US 7,813,482 B2
• Time in Patent Office: Days
• US Class Current: 379/67.100
• CPC Class Codes: H04M 3/436 Arrangements for screening incoming calls , i.e. evaluating the characteristics of a call before deciding whether to answer it based on the calling party profile H04M3/42059; based on location H04M3/42348; based on presence H04M3/42365; diversion H04M3/54
• CPC Class Codes: H04M 3/53333 Message receiving aspects
• CPC Class Codes: H04M 7/006 Networks other than PSTN/ISDN providing telephone service, e.g. Voice over Internet Protocol (VoIP) , including next generation networks with a packet-switched transport layer H04L65/00 takes precedence; aspects not specific to the type of network H04M3/00; special services in those networks H04M3/42

Abstract

A method, system, and program provide for voice mail management. A voice mail filtering controller calculates a separate Bayesian score for each voice mail message from among multiple voice mail message entries received into a voice mailbox for a user, wherein each separate Bayesian score indicates a probability that the associated voice mail message is unwanted by said user. During playback, the voice mail filtering controller automatically deletes a selection of the voice mail messages each with a separate Bayesian score greater than a particular Bayesian score of the last played voice mail message from the voice mailbox.

Claims

1. A method for managing entries within a voice mailbox for at least one user, comprising:

calculating a separate Bayesian score for each voice mail message from among a plurality of voice mail message entries received into said voice mailbox for said user, wherein each said separate Bayesian score indicates a probability that each said separate voice mail message is unwanted by said user; and

responsive to a user selection during playback to delete entries, automatically deleting a selection of said plurality of voice mail message entries from said voice mailbox each with said separate Bayesian score greater than a particular Bayesian score of the last played voice mail message from said voice mailbox.

2. The method according to claim 1, further comprising:

sorting said plurality of voice mail messages entries in an order from lowest Bayesian score to highest Bayesian score; and

playing said selection of said plurality of voice mail messages entries in said order to said user.

3. The method according to claim 1, further comprising:

responsive to calculating said separate Bayesian score for each said voice mail message, automatically deleting a second selection from among said plurality of voice mail messages entries each with said separate Bayesian score greater than a particular score threshold.

4. The method according to claim 1, wherein calculating a separate Bayesian score further comprises:

calculating each said separate Bayesian score from matching words in each said voice mail message to a separate selection of words from a plurality of words within a corpus, wherein each of said plurality of words within said corpus comprises a probability of occurring in a voice mail and a probability of occurring in a voice mail that is unwanted.

5. The method according to claim 4, further comprising:

adding said plurality of words to said corpus from at least one from among a first selection of words detected within at least one voice mail saved by said user during playback, a second selection of words detected during at least one conversation between said user and at least other participant, and a third selection of words detected within at least one voice mail deleted by said user and marked as unwanted voice mail.

6. The method according to claim 1, further comprising:

detecting whether a separate voice characteristic for each said voice mail message from among said plurality of voice mail message entries matches at least one voice imprint from among a plurality of voice imprints, wherein each of said plurality of voice imprints is marked as one from among permitted or excluded; and

automatically deleting a second selection of said plurality of voice mail message entries from said voice mailbox each with said separate voice characteristic matching one from among said plurality of voice imprints marked as excluded.

7. The method according to claim 6, further comprising:

sampling at least a portion of a conversation between said user and at least one other call participant; and

adding said sampling to said plurality of voice imprints marked as excluded.

8. The method according to claim 6, further comprising:

sampling at least a portion of a voice mailbox greeting; and

adding said sampling to said plurality of voice imprints marked as permitted.

9. The method according to claim 6, further comprising:

responsive to detecting a selection by said user to delete a particular voice mail message from among said plurality of voice mail message entries because said particular voice mail message is unwanted by said particular user, adding a sampling of said voice mail message to said plurality of voice imprints marked as excluded.

10. A system for managing entries within a voice mailbox for at least one user, comprising:

a voice mailbox for storing a plurality of voice mail message entries for said user;

said voice mailbox further comprising;

means for calculating a separate Bayesian score for each voice mail message from among said plurality of voice mail message entries stored in said voice mailbox, wherein each said separate Bayesian score indicates a probability that each said separate voice mail message is unwanted by said user; and

means, responsive to a user selection during playback to delete entries, for automatically deleting a selection of said plurality of voice mail message entries from said voice mailbox each with said separate Bayesian score greater than a particular Bayesian score of the last played voice mail message from said voice mailbox.

11. The system according to claim 10, said voice mailbox further comprising:

means for sorting said plurality of voice mail messages entries in an order from lowest Bayesian score to highest Bayesian score; and

means for playing said selection of said plurality of voice mail messages entries in said order to said user.

12. The system according to claim 10, said voice mailbox further comprising:

means, responsive to calculating said separate Bayesian score for each said voice mail message, for automatically deleting a second selection from among said plurality of voice mail messages entries each with said separate Bayesian score greater than a particular score threshold.

13. The system according to claim 10, wherein said means for calculating a separate Bayesian score further comprises:

means for calculating each said separate Bayesian score from matching words in each said voice mail message to a separate selection of words from a plurality of words within a corpus, wherein each of said plurality of words within said corpus comprises a probability of occurring in a voice mail and a probability of occurring in a voice mail that is unwanted.

14. The system according to claim 13, said voice mailbox further comprising:

means for adding said plurality of words to said corpus from at least one from among a first selection of words detected within at least one voice mail saved by said user during playback, a second selection of words detected during at least one conversation between said user and at least other participant, and a third selection of words detected within at least one voice mail deleted by said user and marked as unwanted voice mail.

15. The system according to claim 10, said voice mailbox further comprising:

means for detecting whether a separate voice characteristic for each said voice mail message from among said plurality of voice mail message entries matches at least one voice imprint from among a plurality of voice imprints, wherein each of said plurality of voice imprints is marked as one from among permitted or excluded; and

means for automatically deleting a second selection of said plurality of voice mail message entries from said voice mailbox each with said separate voice characteristic matching one from among said plurality of voice imprints marked as excluded.

16. The system according to claim 15, said voice mailbox further comprising:

means for sampling at least a portion of a conversation between said user and at least one other call participant; and

means for adding said sampling to said plurality of voice imprints marked as excluded.

17. The system according to claim 15, said voice mailbox further comprising:

means for sampling at least a portion of a voice mailbox greeting; and

means for adding said sampling to said plurality of voice imprints marked as permitted.

18. The system according to claim 15, said voice mailbox further comprising:

means, responsive to detecting a selection by said user to delete a particular voice mail message from among said plurality of voice mail message entries because said particular voice mail message is unwanted by said particular user, for adding a sampling of said voice mail message to said plurality of voice imprints marked as excluded.

19. The system according to claim 10, wherein at least a selection of said plurality of voice mail message entries are transmitted through a packet switching network using the voice over Internet Protocol.

20. A program for managing entries within a voice mailbox for at least one user, said program embodied in a computer-readable medium, said program comprising computer-executable instructions which cause a computer to perform the steps of:

controlling calculation of a separate Bayesian score for each voice mail message from among a plurality of voice mail message entries received into said voice mailbox for said user, wherein each said separate Bayesian score indicates a probability that each said separate voice mail message is unwanted by said user; and

controlling automatic deletion of a selection of said plurality of voice mail message entries from said voice mailbox each with said separate Bayesian score greater than a particular Bayesian score of the last played voice mail message from said voice mailbox.