Google llc (20240221772). Phrase Extraction for ASR Models simplified abstract
Contents
Phrase Extraction for ASR Models
Organization Name
Inventor(s)
Ehsan Amid of Mountain View CA (US)
Om Dipakbhai Thakkar of Sunnyvale CA (US)
Rajiv Mathews of Sunnyvale CA (US)
Francoise Beaufays of Mountain View CA (US)
Phrase Extraction for ASR Models - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240221772 titled 'Phrase Extraction for ASR Models
Simplified Explanation: The patent application describes a method for extracting phrases from audio data using an ASR model by modifying the audio data to obfuscate a specific phrase, then comparing the predicted transcription with the ground-truth transcription to identify if the phrase was leaked from the training data.
Key Features and Innovation:
- Phrase extraction method for ASR models
- Modification of audio data to obfuscate a particular phrase
- Comparison of predicted transcription with ground-truth transcription
- Detection of leaked phrases from training data
Potential Applications: This technology can be used in speech recognition systems to improve privacy and security by preventing the leakage of sensitive information during training.
Problems Solved: This technology addresses the issue of unintentional leakage of sensitive phrases from training data used to train ASR models.
Benefits:
- Enhanced privacy and security in speech recognition systems
- Prevention of sensitive information leakage
- Improved accuracy of ASR models
Commercial Applications: Potential commercial applications include secure voice assistants, confidential speech-to-text services, and privacy-focused transcription software.
Prior Art: Prior research in the field of privacy-preserving machine learning and secure speech recognition systems may provide insights into similar methods and technologies.
Frequently Updated Research: Stay updated on advancements in privacy-preserving machine learning, secure speech recognition, and data obfuscation techniques.
Questions about Phrase Extraction Technology: 1. How does the method ensure that the obfuscated phrase is not leaked during the training of the ASR model? 2. What are the potential implications of leaked phrases from training data in ASR models?
Original Abstract Submitted
a method of phrase extraction for asr models includes obtaining audio data characterizing an utterance and a corresponding ground-truth transcription of the utterance and modifying the audio data to obfuscate a particular phrase recited in the utterance. the method also includes processing, using a trained asr model, the modified audio data to generate a predicted transcription of the utterance, and determining whether the predicted transcription includes the particular phrase by comparing the predicted transcription of the utterance to the ground-truth transcription of the utterance. when the predicted transcription includes the particular phrase, the method includes generating an output indicating that the trained asr model leaked the particular phrase from a training data set used to train the asr model.
