EMOTION DETECTION IN VOICEMAIL

• US 20140236596A1
• Filed: 02/21/2013
• Published: 08/21/2014
• Est. Priority Date: 02/21/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Martinez, Raquel Sanchez
• Granted Patent: US 9,569,424 B2
• Time in Patent Office: Days
• US Class Current: 704/235
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G10L 15/187 Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 25/63 for estimating an emotional state
• CPC Class Codes: H04M 2201/60 Medium conversion
• CPC Class Codes: H04M 3/5335 Message type or catagory, e.g. priority, indication

Abstract

Methods and apparatus for processing a voicemail message to generate a textual representation of at least a portion of the voicemail message. At least one emotion expressed in the voicemail message is determined by applying at least one emotion classifier to the voicemail message and/or the textual representation. An indication of the determined at least one emotion is provided in a manner associated with the textual representation of the at least a portion of the voicemail message.

Claims

1. A method for use with a voicemail transcription system that processes a voicemail message and generates a textual representation of at least a portion of the voicemail message, the method comprising:

determining at least one emotion expressed in the voicemail message, wherein the determining comprises applying at least one emotion classifier to the voicemail message; and

providing an indication of the determined at least one emotion in a manner associated with the textual representation of the at least a portion of the voicemail message.

2. The method of claim 1, wherein the determining the at least one emotion comprises analyzing the textual representation of the at least a portion of the voicemail message.

3. The method of claim 1, wherein the providing the indication of the determined at least one emotion comprises:

selecting at least one audio output based, at least in part, on the determined at least one emotion; and

associating the at least one audio output with the textual representation of the at least a portion of the voicemail message.

4. The method of claim 1, wherein the providing the indication of the determined at least one emotion comprises modifying the generated textual representation to include the indication of the determined at least one emotion.

5. The method of claim 4, wherein the modifying the generated text comprises associating at least one graphical symbol representing the determined at least one emotion with the generated textual representation.

6. The method of claim 4, wherein the modifying the generated textual representation comprises modifying a display characteristic of the generated text based, at least in part, on the determined at least one emotion, wherein the display characteristic is selected from the group consisting of:

font size, font color, text highlighting, italics, and boldface.

7. The method of claim 6, wherein the modifying a display characteristic of the generated textual representation comprises modifying a portion of the generated textual representation, wherein the portion of the generated textual representation that is modified is identified based, at least in part, on the emotion determination process.

8. The method of claim 1, wherein providing the indication of the determined at least one emotion comprises displaying an icon separated from but associated with the generated textual representation.

9. The method of claim 8, wherein providing the indication of the determined at least one emotion comprises displaying the icon in a particular color, wherein the particular color is determined based, at least in part, on the determined at least one emotion.

15. The computer-readable storage medium of claim 9, wherein the providing the indication of the determined at least one emotion comprises modifying the generated textual representation to include the indication of the determined at least one emotion.

10. A computer-readable storage medium encoded with a plurality of instructions that, when executed by at least one processor, perform a method for use with a voicemail transcription system that processes a voicemail message and generates a textual representation of at least a portion of the voicemail message, the method comprising:

determining at least one emotion expressed in the voicemail message, wherein the determining comprises applying at least one emotion classifier to the voicemail message; and

providing an indication of the determined at least one emotion in a manner associated with the textual representation of the at least a portion of the voicemail message.

11. The computer-readable storage medium of claim 10, wherein the determining the at least one emotion comprises analyzing the textual representation of the at least a portion of the voicemail message.

12. The computer-readable storage medium of claim 11, wherein the analyzing the textual representation of the at least a portion of the voicemail message comprises comparing at least one word in the generated textual representation to a collection of stored words and/or n-grams associated with one or more emotions.

13. The computer-readable storage medium of claim 10, wherein the at least one emotion classifier is selected from a group consisting of:

support vector machines (SVMs), a logistic regression classifier, a Naï

ve Bayesian classifier, a Gaussian mixture model (GMM), and a statistical language model.

14. The computer-readable storage medium of claim 10, wherein the providing the indication of the determined at least one emotion comprises:

selecting at least one audio output based, at least in part, on the determined at least one emotion; and

associating the at least one audio output with the textual representation of the at least a portion of the voicemail message.

16. A computer system for use with a voicemail transcription system that processes a voicemail message and generates a textual representation of at least a portion of the voicemail message, the computer system comprising:

at least one storage device configured to store at least one emotion classifier for determining at least one emotion expressed in the voicemail message; and

at least one processor programmed to;

determine the at least one emotion, wherein the determining comprises applying the at least one emotion classifier to the voicemail message; and

provide an indication of the determined at least one emotion in a manner associated with the textual representation of the at least a portion of the voicemail message.

17. The computer system of claim 16, wherein the determining the at least one emotion comprises analyzing the textual representation of the at least a portion of the voicemail message.

18. The computer system of claim 16, wherein the providing the indication of the determined at least one emotion comprises:

selecting at least one audio output based, at least in part, on the determined at least one emotion; and

associating the at least one audio output with the textual representation of the at least a portion of the voicemail message.

19. The computer system of claim 16, wherein the providing the indication of the determined at least one emotion comprises modifying the generated textual representation to include the indication of the determined at least one emotion.

20. The computer system of claim 19, wherein the modifying the generated textual representation comprises modifying a display characteristic of the generated text based, at least in part, on the determined at least one emotion, wherein the display characteristic is selected from the group consisting of:

font size, font color, text highlighting, italics, and boldface.