20240013790. METHOD AND SYSTEM OF DETECTING AND IMPROVING REAL-TIME MISPRONUNCIATION OF WORDS simplified abstract (Microsoft Technology Licensing, LLC)

METHOD AND SYSTEM OF DETECTING AND IMPROVING REAL-TIME MISPRONUNCIATION OF WORDS

Organization Name

Microsoft Technology Licensing, LLC

Inventor(s)

Runnan Li of Beijing (CN)

Sheng Zhao of Beijing (CN)

Amit Srivastava of San Jose CA (US)

Huakai Liao of Vancouver (CA)

Ana Parra of San Jose CA (US)

Tapan Bohra of San Jose CA (US)

Akshay Mallipeddi of Cupertino CA (US)

Siliang Kang of Redwood City CA (US)

Lisha Ma of Beijing (CN)

Yinhe Wei of Beijing (CN)

METHOD AND SYSTEM OF DETECTING AND IMPROVING REAL-TIME MISPRONUNCIATION OF WORDS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240013790 titled 'METHOD AND SYSTEM OF DETECTING AND IMPROVING REAL-TIME MISPRONUNCIATION OF WORDS

Simplified Explanation

The patent application describes a method and system for improving pronunciation during a speech. Here is a simplified explanation of the abstract:

• The method involves receiving audio data containing a speech.
• Acoustic scoring and language scoring are performed on the speech.
• A pronunciation score is determined for one or more words based on the acoustic and language scoring.
• If the pronunciation score does not meet a threshold score, it is determined that the word is mispronounced.
• The mispronounced word and its pronunciation score are outputted.

Potential applications of this technology:

• Language learning and pronunciation improvement tools.
• Speech recognition systems that can provide feedback on pronunciation.
• Virtual language tutors or language learning apps.

Problems solved by this technology:

• Difficulty in identifying and correcting mispronunciations during a speech.
• Lack of real-time feedback on pronunciation.
• Inefficient language learning methods that do not focus on pronunciation.

Benefits of this technology:

• Enhanced pronunciation skills for language learners.
• Improved speech recognition accuracy.
• Real-time feedback on pronunciation during a speech.

Original Abstract Submitted

a method and system for enhancing pronunciation during a speech, the method including receiving audio data, the audio data including a speech, performing at least one of acoustic scoring and language scoring on the speech, determining a pronunciation score of one or more words of the speech based on the acoustic scoring and the language scoring, determining that the pronunciation score for the word does not satisfy a threshold score, responsive to determining that the pronunciation score does satisfy the threshold score, identifying the word as mispronounced, and responsive to identifying the word as mispronounced, outputting the word and the pronunciation score thereof.