Adaptive voice-text transmission

Patent metadata

Field	Value
Publication number	US-2017078490-A1
Application number	US-201514856428-A
Country	US
Kind code	A1
Filing date	Sep 16, 2015
Priority date	Sep 16, 2015
Publication date	Mar 16, 2017
Grant date	—

Abstract

Responsive to determining to transition a voice call from voice communications over a voice network to streamed text over a packetized data network, a voice conversation correlation identifier is created that identifies the voice call and specifies incoming and outgoing streamed text data as part of the voice call. Additional outgoing speech spoken by a user associated with the voice call is converted to streamed text data. The streamed text data identified by the voice conversation correlation identifier is sent within an outgoing text stream over the packetized data network. Streamed response text data identified by the voice conversation correlation identifier is received within an incoming text stream over the packetized data network. The received streamed response text data within the incoming text stream is converted to speech output as part of the voice call.

First claim

1 . A computer-implemented method, comprising:
creating, responsive to determining to transition a voice call from voice communications over a voice network to streamed text over a packetized data network, a voice conversation correlation identifier that identifies the voice call and specifies incoming and outgoing streamed text data as part of the voice call; converting additional outgoing speech spoken by a user associated with the voice call to streamed text data; sending the streamed text data identified by the voice conversation correlation identifier within an outgoing text stream over the packetized data network; receiving streamed response text data identified by the voice conversation correlation identifier within an incoming text stream over the packetized data network; and converting the received streamed response text data within the incoming text stream to speech output as part of the voice call; where receiving the streamed response text data identified by the voice conversation correlation identifier within the incoming text stream over the packetized data network comprises one of:
receiving text advertisement content within the received streamed response text data within the incoming text stream associated with a subject matter of the voice call and rendering the text advertisement content during the voice call; or
receiving text represented in a different language from a language spoken by the user and performing a language conversion of the text represented in the different language to the language spoken by the user, where the received streamed response text data converted to the speech output comprises the language-converted text.

2 . The computer-implemented method of claim 1 , where:
determining to transition the voice call from voice communications over the voice network to the streamed text over the packetized data network comprises detecting a user request from the user to initiate language translation to the different language relative to the language spoken by the user during the voice call; converting additional outgoing speech spoken by the user associated with the voice call to the streamed text data further comprises translating the streamed text data to the different language; and sending the streamed text data identified by the voice conversation correlation identifier within the outgoing text stream over the packetized data network comprises sending the translated different language streamed text data identified by the voice conversation correlation identifier within the outgoing text stream over the packetized data network.

3 . The computer-implemented method of claim 1 , where determining to transition the voice call from the voice network to the streamed text over the data network comprises:
detecting signal degradation of the voice network.

4 . The computer-implemented method of claim 1 , where determining to transition the voice call from voice communications over the voice network to the streamed text over the packetized data network comprises:
determining to insert an outgoing advertisement into the voice call; and inserting outgoing text advertisement content of the outgoing advertisement into the outgoing streamed text data.

5 . The computer-implemented method of claim 1 , further comprising:
analyzing a subject matter topic of the voice call; selecting advertisement content that corresponds to the subject matter topic of the voice call; and inserting outgoing text advertisement content of the selected advertisement content into the outgoing streamed text data.

6 . (canceled)

7 . The computer-implemented method of claim 1 , further comprising:
switching from the streamed text over the packetized data network back to the voice communications over the voice network in response to determining that signal strength on an available voice network is sufficient to utilize the voice communications for the voice call.

8 . A system, comprising:
a communication device; and a processor programmed to:
create, responsive to determining to transition a voice call from voice communications over a voice network to streamed text over a packetized data network, a voice conversation correlation identifier that identifies the voice call and specifies incoming and outgoing streamed text data as part of the voice call;
convert additional outgoing speech spoken by a user associated with the voice call to streamed text data;
send, via the communication device, the streamed text data identified by the voice conversation correlation identifier within an outgoing text stream over the packetized data network;
receive, via the communication device, streamed response text data identified by the voice conversation correlation identifier within an incoming text stream over the packetized data network; and
convert the received streamed response text data within the incoming text stream to speech output as part of the voice call;
where, in being programmed to receive, via the communication device, the streamed response text data identified by the voice conversation correlation identifier within the incoming text stream over the packetized data network the processor is programmed to one of:
receive text advertisement content within the received streamed response text data within the incoming text stream associated with a subject matter of the voice call and rendering the text advertisement content during the voice call; or
receive text represented in a different language from a language spoken by the user and performing a language conversion of the text represented in the different language to the language spoken by the user, where the received streamed response text data converted to the speech output comprises the language-converted text.

9 . The system of claim 8 , where:
in being programmed to determine to transition the voice call from voice communications over the voice network to the streamed text over the packetized data network, the processor is programmed to detect a user request from the user to initiate language translation to the different language relative to the language spoken by the user during the voice call; in being programmed to convert additional outgoing speech spoken by the user associated with the voice call to the streamed text data, the processor is further programmed to translate the streamed text data to the different language; and in being programmed to send, via the communication device, the streamed text data identified by the voice conversation correlation identifier within the outgoing text stream over the packetized data network, the processor is programmed to send the translated different language streamed text data identified by the voice conversation correlation identifier within the outgoing text stream over the packetized data network.

10 . The system of claim 8 , where, in being programmed to determine to transition the voice call from the voice network to the streamed text over the data network, the processor is programmed to one of:
detect signal degradation of the voice network; or determine to insert an outgoing advertisement into the voice call and insert outgoing text advertisement content of the outgoing advertisement into the outgoing streamed text data.

11 . The system of claim 8 , where the processor is further programmed to:
analyze a subject matter topic of the voice call; select advertisement content that corresponds to the subject matter topic of the voice call; and insert outgoing text advertisement content of the selected advertisement content into the outgoing streamed text data.

12 . (canceled)

13 . The system of claim 8 , where the

Assignees

• IBM

Inventors

• Kaminsky David L
• Ogle David M

Classifications

• H04M3/4878

  Advertisement messages · CPC title

• G06F40/58

  Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation · CPC title

• G06F40/40

  Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30) · CPC title

• H04M2201/60

  Medium conversion · CPC title

• H04M2201/39

  using speech synthesis · CPC title