Using a speech analytics system to control whisper audio

• US 9,674,357 B1
• Filed: 02/18/2014
• Issued: 06/06/2017
• Est. Priority Date: 07/24/2013
• Status: Active Grant

• Current Assignee: Alvaria Cayman
• Original Assignee: Noble Systems Corporation
• Inventors: Pycko, Marcin, Koster, Karl H.
• Primary Examiner(s): Hoque, Nafiz E
• Application Number:
  US14/183,083
• Time in Patent Office: 1,204 Days
• Field of Search: 37926501, 37926506, 37926507, 37926509
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2203/306 Prerecordings to be used during a voice call
• CPC Class Codes: H04M 2203/5018 Initiating a conference during a two-party conversation, i.e. three-party-service or three-way-call computer conference organisation arrangements H04L12/1818
• CPC Class Codes: H04M 2203/552 Call annotations
• CPC Class Codes: H04M 3/00 Automatic or semi-automatic exchanges
• CPC Class Codes: H04M 3/428 Arrangements for placing incoming calls on hold
• CPC Class Codes: H04M 3/5175 Call or contact centers supervision arrangements

Abstract

A speech analytics system monitors speech from a remote party and/or agent to determine when a whisper is to be provided to the agent. A whisper is aural information that may assist the agent at certain opportune times. The whisper may be verbal or non-verbal in nature. A whisper bridge is used that allows the whisper to be heard by the agent, but not by the remote party. In one embodiment, the speech analytics system may cause visual-based agent information to be provided to the computer used by the agent, and monitor the agent'"'"'s speech to ascertain if an expected speech response to the whisper was provided by the agent. If not, the speech analytics system may cause the whisper to be provided to the agent, as a form of reminder.

Claims

1. A method for providing a whisper to an agent using a whisper bridge, comprising:

establishing a first call leg between the whisper bridge and a remote party;

establishing a second call leg between the whisper bridge and the agent;

establishing a third call leg between the whisper bridge and a speech analytics system (“

SAS”

), wherein the SAS comprises an Agent Whisper Control Application Specific Module (“

AWC-ASM”

) comprising a processor, and wherein the third call leg comprises a first voice channel and a second voice channel conveying speech from the agent and remote party respectively;

monitoring speech of the third call leg by the SAS;

detecting a keyword in the speech of the third call leg;

providing a speech-related event notification to the AWC-ASM associated with the keyword detected in the speech, the speech-related event notification comprising a call identifier identifying the third call leg and an identifier indicating one of the agent or the remote party as originating the keyword; and

causing the whisper to be provided to the agent, wherein the whisper bridge prevents the remote party from hearing the whisper.

2. The method of claim 1, further comprising:

providing a message from the AWC-ASM to an announcement system in response to receiving the speech-related event notification, the message identifying a pre-recorded audio file that is to be played to the agent as the whisper.

3. The method of claim 1, wherein causing the whisper to be provided to the agent comprises causing a text-to-speech generator in the SAS to generate the whisper.

4. The method of claim 1, further comprising:

providing visual information to a computer used by the agent in response to detecting the keyword in the speech by the SAS.

5. The method of claim 4, wherein the whisper comprises a non-verbal sound.

6. The method of claim 1, wherein the whisper bridge comprises a whisper conference bridge and the method further comprises:

causing another person to be bridged onto the whisper conference bridge, wherein audio from the another person is provided to a phone device used by the agent and the audio from the another person is not heard by the remote party.

7. The method of claim 1, wherein the whisper bridge is configured to process Voice over IP (“

VoIP”

) audio data and comprises structure partly located in a call handler and in an agent'"'"'s computer.

8. The method of claim 7, wherein causing the whisper to be provided to the agent comprises providing the whisper to the agent'"'"'s computer comprising a soft-phone used by the agent.

9. The method of claim 1, further comprising:

ascertaining one of a plurality of whispers that the AWC-ASM is configured to provide based on receiving one of a plurality of speech-related event notifications.

10. A system for providing a whisper to an agent in a contact center, comprising:

a speech analytics system (“

SAS”

) configured to process a first call leg, wherein the first call leg conveys audio of a call involving the agent and a remote party, the SAS configured to generate a message identifying an audio informational message to be provided to the agent;

a whisper bridge configured to bridge the first call leg with the SAS, a second call leg with the remote party, a third call leg with the agent, and a fourth call leg, wherein the audio of the fourth call leg is not provided to the second call leg of the remote party; and

an announcement component configured to receive the message from the SAS and in response, provide the audio informational message to the whisper bridge on the fourth call leg, wherein the audio informational message is provided to the third call leg with the agent but not to the second call leg with the remote party.

11. The system of claim 10, wherein the announcement component comprises a text-to-speech generator for generating the audio information message in response to receiving the message.

12. The system of claim 10, wherein the first call leg with the SAS comprises a first audio channel conveying speech from the remote party and a second audio channel conveying speech from the agent.

13. The system of claim 10,

wherein the SAS comprises a first processor configured to:

detect a keyword in the audio of the call, and

provide a speech-related event notification reflecting detection of the keyword; and

wherein the SAS comprises a second processor configured to;

receive the speech-related event notification reflecting detection of the keyword, and

generate the message to the announcement component in response to receiving the speech-related event notification.

14. The system of claim 13, wherein audio of the second call leg and the third call leg is not provided to the announcement component on the fourth call leg.

15. The system of claim 13, wherein the whisper bridge is distributed between a call handler and a computer used by the agent.