Real-time transcription and interaction with a caller based on the transcription

• US 9,736,309 B1
• Filed: 08/19/2016
• Issued: 08/15/2017
• Est. Priority Date: 08/19/2016
• Status: Expired due to Fees

• Current Assignee: Circle River, Inc.
• Original Assignee: Circle River, Inc.
• Inventors: Bentitou, Adam, Mansfield, David, Lippman, Robert
• Primary Examiner(s): Huynh, Van D
• Application Number:
  US15/241,555
• Time in Patent Office: 361 Days
• Field of Search: 379 8813- 8814
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H04M 2201/39 using speech synthesis
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 2201/60 Medium conversion
• CPC Class Codes: H04M 3/527 Centralised call answering arrangements not requiring operator intervention

Abstract

Receiving a telephone call to an auto-attendant, artificial intelligence, or person takes place. While this phone call is being conducted, a speech to text transcription is created and sent in real-time to another person at another network node. This person can read the transcript and interact with the phone call by sending his or her own commands, text, or speech to be made part of the phone call.

Claims

1. A method of communicating with a calling party using artificial intelligence when receiving a telephone call, comprising the steps of:

receiving a phone call at a first network node from a calling party;

based on speech recognition of audio received from said calling party, creating a transcription of said audio received from said calling party;

transmitting audio output of at least one of;

audio based on text to speech synthesis; and

pre-recorded audio to said calling party of said telephone call;

said audio output comprising responses based on said speech recognition; and

while creating said transcription of at least some of said audio of said incoming telephone call, sending said transcription to a bidirectional transceiver at a second network node in real-time, wherein sending said transcription to said bidirectional transceiver causes a plurality of selectable elements to be exhibited on said bidirectional transceiver, wherein said selectable elements are based on preceding conversations between said calling party and said artificial intelligence;

wherein the artificial intelligence communicates with the calling party automatically and independently unless overridden by instructions from the bidirectional transceiver.

2. The method of claim 1, wherein said audio output is based partially on artificial intelligence and partially on instructions received from said bidirectional transceiver receiving said transcription.

3. The method of claim 2, further comprising the step of receiving text input from said bidirectional transceiver;

and

playing a speech synthesized version of said text input as part of said audio output.

4. The method of claim 2, further comprising the step of receiving audio input from said bidirectional transceiver;

converting said audio input to text; and

playing a speech synthesized version of said audio input, based on said text, such that said speech synthesized version matches a voice of said speech synthesis in said audio output.

5. The method of claim 4, further comprising repeating said step of transmitting audio output of said speech synthesis comprising said responses based on said speech recognition, after said step of playing said speech synthesized version of said audio input.

6. The method of claim 1, wherein a selectable element of said selectable elements comprises at least one selector which, when selected, causes said call to be forwarded to another network node or called party.

7. The method of claim 1, wherein a selectable element of said selectable elements comprises at least one selector, which, when selected, causes future calls from said calling party received at said first network node, to be forwarded to said bidirectional transceiver, bypassing said step of said creating said transcription.

8. The method of claim 1, wherein a selectable element of said selectable elements comprises at least one selector, which, when selected causes future calls from said calling party to carry out said step of said receiving said phone call at said first network node and said using speech recognition while skipping or suppressing said step of sending said transcription to said bidirectional transceiver.

9. The method of claim 1, wherein said preceding conversation is detected as being related to scheduling a meeting between said calling party and another party, and said selectable elements comprise selections related to time.

10. The method of claim 1, wherein said transcription further comprises a transcription of said text which is part of said transmitting of said audio output.

11. A network switch, comprising:

at least one phone network interface which receives phone calls from a calling party at a first network node;

a physical storage medium which stores audio from said phone calls;

a speech recognition engine which transcribes at least some of said audio from said phone calls;

a transcription engine which transcribes at least some of said audio from said phone calls;

a packet-switched data network connection which transmits audio output of at least one of;

audio based on text to speech synthesis; and

pre-recorded audio to said calling party of said telephone call;

wherein said audio output comprises responses based on output of transcriptions from said speech recognition engine or said transcription engine; and

while transcribing said at least some of said audio of said telephone call, sending said transcription to a bidirectional transceiver at a second network node in real-time, wherein sending said transcription to said bidirectional transceiver causes a plurality of selectable elements to be exhibited on said bidirectional transceiver, wherein said selectable elements are based on preceding conversations between said calling party and said artificial intelligence,

wherein the artificial intelligence communicates with the calling party automatically and independently unless overridden by instructions from the bidirectional transceiver.

12. The network switch of claim 11, wherein said audio output is based partially on artificial intelligence and partially on instructions received from said bidirectional transceiver receiving said transcription.

13. The network switch of claim 12, wherein data are transmitted via said packet-switched data network to said bidirectional transceiver causing the plurality of selectable elements to be exhibited on said bidirectional transceiver, wherein said selectable elements are based on preceding conversation between said calling party and said artificial intelligence.

14. The network switch of claim 12, wherein text input from said bidirectional transceiver is received via said packet-switched data network connection;

and

playing a speech synthesized version of said text input as part of said audio output.

15. The network switch of claim 12, wherein audio input from said bidirectional transceiver is received via said packet-switched data network connection;

said audio input is converted to text; and

a speech synthesized version of said audio input, based on said text, is exhibited over said phone network, such that said speech synthesized version matches a voice of said speech synthesis in said audio output.

16. The network switch of claim 13, wherein a selectable element of said selectable elements comprises at least one selector which, when selected, causes said call to be forwarded to another network node or called party.

17. The network switch of claim 16, wherein a selectable element of said selectable elements comprises at least one selector which, when selected, causes future calls from said calling party received at said first network node to be forwarded to said bidirectional transceiver, bypassing said step of said creating said transcription.

18. The network switch of claim 13, wherein a selectable element of said selectable elements comprises at least one selector, which, when selected, causes future calls from said calling party to carry out said step of said receiving said phone call at said first network node and said using speech recognition while skipping or suppressing said step of sending said transcription to said bidirectional transceiver.

19. The network switch of claim 13, wherein said preceding conversation is detected as being related to scheduling a meeting between said calling party and another party, and said selectable elements comprise selections related to time.