Translation with conversational overlap

Patent metadata

Field	Value
Publication number	US-10437934-B2
Application number	US-201715650561-A
Country	US
Kind code	B2
Filing date	Jul 14, 2017
Priority date	Sep 27, 2016
Publication date	Oct 8, 2019
Grant date	Oct 8, 2019

Abstract

A plurality of utterances of a first user from the language of the first user is translated into a language of a second user. The confidence scores associated with the translated utterances are compared with a confidence threshold. A predetermined utterance gap is adjusted based on the comparison. The predetermined utterance gap is a duration of time that occurs between utterances.

First claim

What is claimed is:

1. A method, comprising:
receiving a plurality of utterances of a first person in a first language;
detecting an utterance gap between sequential utterances of the plurality of utterances;
determining, prior to translating an utterance, whether the utterance will be translated by comparing the utterance gap after the utterance is completed to a threshold utterance gap and, if it is determined that the utterance will be translated:
translating the utterance from the first language to a second language to produce a translated utterance;
determining a translation confidence score for the translated utterance;
determining whether the confidence score is greater than or equal to a confidence level;
determining, based on whether the confidence score is greater than or equal to the confidence level, whether the confidence score is great enough to output the translated utterance; and
determining accrued translation confidence scores for a plurality of utterances, wherein the threshold utterance gap is increased if a percentage of the accrued translation confidence scores is less than a confidence threshold.

2. The method of claim 1 , wherein the plurality of utterances include a first utterance of the first person and a second utterance of the first person, wherein the first utterance of the first person and the second utterance of the first person are received at a device associated with second person.

3. The method of claim 2 , wherein the first utterance of the first person and the second utterance of the first person are data associated with spoken utterances transmitted from a device associated with the first person.

4. The method of claim 2 , wherein the first utterance of the first person and the second utterance of the first person are spoken utterances of the first person.

5. The method of claim 1 , wherein the threshold utterance gap is less than a turn threshold duration.

6. The method of claim 1 , further comprising outputting the translated utterance at a device associated with a second person.

7. The method of claim 6 , wherein the device associated with the second person includes a pair of earbuds.

8. The method of claim 7 , wherein the pair of earbuds are configured to occlude a direct sound path associated with the plurality of utterances of the first person by attenuating the plurality of utterances.

9. The method of claim 8 , wherein an amount of attenuation of the plurality of utterances is adjustable.

10. The method of claim 6 , wherein the translated utterance is outputted to appear to come from a predetermined spatial location.

11. The method of claim 6 , wherein the translated utterance is outputted to appear to come from a spatial location of the first person.

12. The method of claim 1 , wherein the threshold utterance gap is adjustable based at least in part on a speech pattern of the first person.

13. The method of claim 1 , wherein the threshold utterance gap is adjustable based at least in part on a cadence of the first person's speech.

14. A system, comprising:
a processor configured for:
receiving a plurality of utterances of a first person in a first language;
detecting an utterance gap between sequential utterances of the plurality of utterances;
determining, prior to translating an utterance, whether the utterance will be translated by comparing the utterance gap after the utterance is completed to a threshold utterance gap and, if the processor determines that the utterance will be translated:
translating the utterance from the first language to a second language to produce a translated utterance;
determining a translation confidence score for the translated utterance;
determining whether the confidence score is greater than or equal to a confidence level;
determining, based on whether the confidence score is greater than or equal to the confidence level, whether the confidence score is great enough to output the translated utterance; and
determining accrued translation confidence scores for a plurality of utterances, wherein the threshold utterance gap is decreased if a percentage of the accrued translation confidence scores is greater than or equal to a confidence threshold and wherein the confidence threshold corresponds with a percentage of translations that are accurate.

15. The system of claim 14 , wherein the threshold utterance gap is increased if a percentage of the accrued translation confidence scores is less than the confidence threshold.

16. The system of claim 14 , wherein the processor is further configured to output the translated plurality of utterances at a device associated with a second person.

17. A computer program product, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
receiving a plurality of utterances of a first person in a first language;
detecting an utterance gap between sequential utterances of the plurality of utterances;
determining, prior to translating an utterance, whether the utterance will be translated by comparing the utterance gap after the utterance is completed to a threshold utterance gap and, if it is determined that the utterance will be translated:
translating the utterance from the first language to a second language to produce a translated utterance;
determining a translation confidence score for the translated utterance;
determining whether the confidence score is greater than or equal to a confidence level;
determining, based on whether the confidence score is greater than or equal to the confidence level, whether the confidence score is great enough to output the translated utterance; and
determining accrued translation confidence scores for a plurality of utterances, wherein the threshold utterance gap is increased if a percentage of the accrued translation confidence scores is less than a confidence threshold.

18. The system of claim 14 , wherein the processor is further configured for determining whether a maximum condition is satisfied if the processor determines that the confidence score is not great enough to output the translated utterance.

19. The system of claim 18 , wherein the processor is further configured for combining the translated utterance with a subsequent utterance if the processor determines that the maximum condition is not satisfied.

Assignees

• Dolby Laboratories Licensing Corp

Inventors

• Baker Jeffrey
• Garcia Sal Gregory
• Long Paul Anthony

Classifications

• G06F40/263 Primary

  Language identification · 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

• G06F40/47 Primary

  Machine-assisted translation, e.g. using translation memory · CPC title

• H04R1/1083

  Reduction of ambient noise (active noise reduction per se G10K11/175; protective devices for the ear, e.g. providing acoustic protection A61F11/06) · CPC title

• H04R3/005

  for combining the signals of two or more microphones (specially adapted for hearing aids H04R25/407) · CPC title