20240028843. CROSS-LINGUAL VOICE CONVERSION SYSTEM AND METHOD simplified abstract (TMRW Foundation IP S. À R.L.)
Contents
CROSS-LINGUAL VOICE CONVERSION SYSTEM AND METHOD
Organization Name
Inventor(s)
CROSS-LINGUAL VOICE CONVERSION SYSTEM AND METHOD - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240028843 titled 'CROSS-LINGUAL VOICE CONVERSION SYSTEM AND METHOD
Simplified Explanation
The abstract describes a cross-lingual voice conversion system and method. Here are the bullet points explaining the patent/innovation:
- The system extracts audio features from voice audio segments in different languages.
- The extracted features include speaker-dependent acoustic features from the first language and speaker-independent linguistic features from the second language.
- Generators receive the extracted features and produce a third voice candidate that retains the acoustic features from the first language and the linguistic features from the second language.
- Discriminators compare the third voice candidate with ground truth data and provide feedback to the generators for refining the candidate.
- The system aims to convert a voice from one language to another while preserving the speaker's voice characteristics.
Potential applications of this technology:
- Language learning and pronunciation improvement: The system can be used to convert a learner's voice in their native language to sound like a native speaker in the target language, aiding in language acquisition and pronunciation practice.
- Dubbing and voice-over: The system can be utilized in the entertainment industry to convert voices from one language to another for dubbing movies, TV shows, or voice-over work.
- Accessibility: The technology can assist individuals with language barriers by converting speech from one language to another in real-time, enabling better communication and understanding.
Problems solved by this technology:
- Language barrier: The system addresses the challenge of language barriers by providing a means to convert voices from one language to another, facilitating communication between individuals who speak different languages.
- Voice conversion accuracy: By incorporating both speaker-dependent acoustic features and speaker-independent linguistic features, the system aims to improve the accuracy and naturalness of the converted voice.
Benefits of this technology:
- Multilingual communication: The system enables effective communication between individuals speaking different languages, promoting cultural exchange and understanding.
- Personalized voice conversion: By retaining the speaker's voice characteristics, the technology allows for personalized voice conversion, maintaining the individual's identity while speaking in a different language.
- Enhanced language learning: The system can assist language learners in improving their pronunciation and fluency by providing them with a voice that closely resembles that of a native speaker.
Original Abstract Submitted
a cross-lingual voice conversion system and method comprises a voice feature extractor configured to receive a first voice audio segment in a first language and a second voice audio segment in a second language, and extract, respectively, audio features comprising first-voice, speaker-dependent acoustic features and second-voice, speaker-independent linguistic features. one or more generators are configured to receive extracted features, and produce therefrom a third voice candidate keeping the first-voice, speaker-dependent acoustic features and the second-voice, speaker-independent linguistic features, wherein the third voice candidate speaks the second language. one or more discriminators are configured to compare the third voice candidate with the ground truth data, and provide results of the comparison back to the generator for refining the third voice candidate.
