Text-to-Voice Machine Translation

Summary:
Write in one language, see and hear, immediately, the translation in another language. For example, write in German, see and hear the translation in English,, immediately, without needing another, external program to read aloud the translation.

Full Description:
Write in one language, see and hear, immediately, the translation in another language. For example, write in German, see and hear the translation in English. Our text to speech technology is based on whole words and concepts, not on phonemes and syllables, as that of the competition. For example, the text to voice program will pronounce "fish" if one writes "herring", etc. Many individual words, such as prepositions, articles, etc. are pronounced as individual words. Many other individual words, such as names of important places or people are also pronounced as individual words. The accent and the stress of the spoken word may depend on whether the word is used as an adjective or as a verb or other part of speech. For example, such words as "absent", “access”, “accent”, etc. have different stress when used as an adjective noun or verb, in context. There are many languages in the world, for example Vietnamese, with many homographs, where the pronunciation of the word is very important because different pronunciations of the same word have different meanings.
Our program recognizes the part of speech of the word, in context, and chooses the correct pronunciation. The existing text-to-voice software programs cannot do that, because they do not analyse the sentence to determine the role of each word, in context.
As such, these programs can be regarded as educational, with inherent possibility to use them in other spheres. Patent pending.

Problem this idea/invention addresses:
Helps language acquisition and solves communication problems between users not speaking the same language.
There are text to voice programs on the market, but there are no such programs that recognize the meaning of the word and work on the level of the concept. Besides, the programs of the competition can pronounce any nonsense you write and they need an external program to read aloud the translation. Our programs pronounce only words or meanings that exist in the language and do that immediately. As such, our programs fully preserve the accent of the spoken word, while those of the competition cannot do that.

Auxiliary products or services for sale:
Natural Language (text) Processing software development technology: parsing, spell-checking (grammatical and orthographical), machine translation (incl. into plain English and plain German), content recognition, question answering, logical inference, text to voice, text to video, text summarisation and text attribution for English, German, French, Italian. Asking price: we will consider an offer.

Attached files:
d-e-tr-vo.jpg

Asking price: [CONTACT SELLER]
Available for consultation? Yes

Invention #11755
Date posted: 2012-05-13


« More Software Inventions


Share on