Method and apparatus for cross-lingual communication

• US 9,552,354 B1
• Filed: 03/12/2013
• Issued: 01/24/2017
• Est. Priority Date: 09/05/2003
• Status: Active Grant

• Current Assignee: ZAMA Innovations LLC
• Original Assignee: Spoken Translation, Inc.
• Inventors: Seligman, Mark, Dillinger, Mike
• Primary Examiner(s): Yang, Qian
• Application Number:
  US13/797,628
• Time in Patent Office: 1,414 Days
• Field of Search: 704/2, 704/3
• US Class Current: 1/1
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/51 Translation evaluation
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A system and method for a highly interactive style of speech-to-speech translation is provided. The interactive procedures enable a user to recognize, and if necessary correct, errors in both speech recognition and translation, thus providing robust translation output than would otherwise be possible. The interactive techniques for monitoring and correcting word ambiguity errors during automatic translation, search, or other natural language processing tasks depend upon the correlation of Meaning Cues and their alignment with, or mapping into, the word senses of third party lexical resources, such as those of a machine translation or search lexicon. This correlation and mapping can be carried out through the creation and use of a database of Meaning Cues, i.e., SELECT. Embodiments described above permit the intelligent building and application of this database, which can be viewed as an interlingua, or language-neutral set of meaning symbols, applicable for many purposes. Innovative techniques for interactive correction of server-based speech recognition are also described.

Claims

1. A system providing cross-linguistic communication and providing feedback for machine learning, comprising:

a client component capturing speech and tactile inputs, the client component providing a user interface configured to display and correct inputs and meanings for the inputs, a translation of an input term into a different language than an input language, a retranslation from the target language back into the source language, and interface elements indicating one of whether translations should be verified before transmission, whether a previous translation should be revised, or whether pre-translated phrases are available for execution; and

a server component providing the meanings, the translation and the retranslation to the client component based upon the input term, the server component including,

an interaction manager stored within memory of the server component, the interaction manger configured to request the translation of the inputs and access a database of Meaning Cues corresponding to the inputs, the corresponding Meaning Cues representing a sense of the input term, the sense including a synonym for each different sense of the input term, wherein each different sense includes a different meaning for the term.

2. The system of claim 1 wherein the Meaning Cues are derived through one of rule-based techniques, statistically based techniques, or some combination of rule based and statistically based techniques.

3. The system of claim 2, wherein Meaning Cues include synonym sets derived from the translation phrase tables of a given statistical machine translation system.

4. The system of claim 3, wherein the synonym sets are derived from the translation phrase tables of a secondary machine translation system which serves as a proxy for the primary translation system.

5. The system of claim 1 further comprising:

a memory database storing user profiles defining word sense preferences for each user profile.

6. The system of claim 1, wherein the server component includes a speech recognition module in communication with the interaction manager, the speech recognition module permitting use of voice commands captured through the client component.

7. The system of claim 1, wherein the server component includes a text-to-speech module providing an audio signal to the client component for pronunciation of the input term in the different language and the input language.