Dynamic translation for a conversation

Patent metadata

Field	Value
Publication number	US-11908450-B2
Application number	US-202016883768-A
Country	US
Kind code	B2
Filing date	May 26, 2020
Priority date	May 26, 2020
Publication date	Feb 20, 2024
Grant date	Feb 20, 2024

Abstract

A conversation design is received for a conversation bot that enables the conversation bot to provide a service using a conversation flow specified at least in part by the conversation design. The conversation design specifies in a first human language at least a portion of a message content to be provided by the conversation bot. It is identified that an end-user of the conversation bot prefers to converse in a second human language different from the first human language. In response to a determination that the message content is to be provided by the conversation bot to the end-user, the message content of the conversation design is dynamically translated for the end-user from the first human language to the second human language. The translated message content is provided to the end-user in a message from the conversation bot.

First claim

What is claimed is:

1. A method comprising:
receiving, for a conversation bot, a conversation design that enables the conversation bot to provide a service using a conversation flow specified at least in part by the conversation design, wherein the conversation design specifies in a first human language at least a portion of a message content to be provided by the conversation bot;
deploying the conversation design with the message content in the first human language;
identifying that an end-user of the conversation bot prefers to converse in a second human language different from the first human language;
in response to a determination that the message content is to be provided by the conversation bot to the end-user, dynamically translating for the end-user the message content of the conversation design from the first human language to the second human language, wherein the dynamic translation from the first human language to the second human language is performed using a machine translator further in response to deploying the conversation design;
providing to the end-user the translated message content in a message from the conversation bot, wherein providing the translated message content includes combining the translated message content in the second human language and the message content in the first human language as a single combined conversation bot message presented visually together in the first human language and the second human language as from the same conversation bot;
determining to transfer the end-user from the conversation bot to a first live agent;
adding a second live agent to an existing conversation between the end-user and the first live agent; and
receiving a new message from the second live agent in a fourth human language, translating the new message from the fourth human language to the second human language and a third human language, providing the end-user the translation of the new message in the second human language, and providing the first live agent the translation of the new message in the third human language.

2. The method of claim 1 , wherein the conversation design is deployed with the message content only prespecified in the first human language.

3. The method of claim 1 , wherein the message from the conversation bot is provided to the end-user as a text message.

4. The method of claim 1 , wherein the message from the conversation bot is provided to the end-user as a voice synthesized audio message.

5. The method of claim 4 , further comprising, in response to an indication from the end-user, stopping an audio output of the voice synthesized audio message prior to completion of the audio output of the entire voice synthesized audio message.

6. The method of claim 1 , wherein the message from the conversation bot includes a plurality of response options presented visually within the same message as selectable within the message and each of the response options is provided both in the second human language and in the first human language visually together in a same chat message bubble from the conversation bot.

7. The method of claim 1 , further comprising:
in response to a determination that the message content is to be provided by the conversation bot to a second end-user associated with a third human language, dynamically translating for the second end-user, the message content of the conversation design from the first human language to the third human language, wherein the dynamic translation from the first human language to the third human language is performed using the machine translator.

8. The method of claim 1 , further comprising at least one of the following:
(a) automatically switching between voice speech modes of: listening mode, speaking mode, and processing wait mode;
(b) providing a live voice speech transcription of at least a portion of a voice speech input while the end-user is providing the voice speech input; or
(c) receiving a voice speech input from the end-user and automatically determining that the voice speech input has ended based on a detection of silence for a dynamically selected threshold amount of time.

9. The method of claim 1 , wherein the message content specifies a first portion to be provided as a text content portion of the message and a second portion to be provided as an audio speech message content portion of the message, translating the message content includes separately translating the first portion from the second portion, and providing the message to the end-user includes providing the translated first portion as a text output and providing the translated second portion as an audio speech output associated with the text output.

10. The method of claim 1 , further comprising providing to the first live agent a translation in the third human language of previous messages between the conversation bot and the end-user.

11. The method of claim 1 , further comprising receiving a second new message from the first live agent in the third human language, translating the second new message from the third human language to the second human language, and providing the end-user the translation of the second new message in the second human language.

12. The method of claim 11 , wherein the translated second new message is provided to the end-user as a voice synthesized audio message using a voice persona that is different from a previous voice persona utilized to provide the message from the conversation bot.

13. The method of claim 1 , further comprising:
receiving a response voice speech input from the end-user in response to a request for a selection among option choice items;
processing identifiers of the option choice items;
determining a corresponding match score for each of the processed identifiers of the option choice items based on a corresponding comparison associated with the received response voice speech input; and
identifying one of the option choice items as a chosen option choice item based on the match scores.

14. The method of claim 13 , wherein the processing identifiers of the option choice items includes performing a translation from the first human language to the second human language.

15. The method of claim 13 , wherein the comparison associated with the received response input allows partial matching for a numerical word of the response input.

16. A system comprising:
one or more processors configured to:
receive, for a conversation bot, a conversation design that enables the conversation bot to provide a service using a conversation flow specified at least in part by the conversation design, wherein the conversation design specifies in a first human language at least a portion of a message content to be provided by the conversation bot;
deploy the conversation design with the message content in the first human language;
identify that an end-user of the conversation bot prefers to converse in a second human language different from the first human language;
in response to a determination that the message content is to be provided by the conversation bot to the end-user, dynamically translate for the end-user, the message content of the conversation design from the first human language to the second human language, wherein the dynamic translation from the first human language to the second human language is performed using a machine translator further in response to deploying the conversation design;
provide to the end-user the translated message content in a message from the conversation bot, wherein providing the translated message content includes combining the translated message

Assignees

• Servicenow Inc

Inventors

• Swvigaradoss Jebakumar Mathuram Santhosm
• Sunkara Satya Sarika
• Goel Ankit
• Voleti Rajesh
• Verma Rishabh
• Casey Patrick
• Surapaneni Rao

Classifications

• G10L13/086 Primary

  Detection of language · CPC title

• G06F40/274

  Converting codes to words; Guess-ahead of partial word inputs · CPC title

• G06F40/58 Primary

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

• G10L15/005

  Language recognition · CPC title

• G10L13/00

  Speech synthesis; Text to speech systems · CPC title