Otavio Braga

Title: Otavio Braga: Innovator in Audio-Visual Speech Recognition

Introduction

Otavio Braga is a prominent inventor based in Mountain View, CA (US), known for his contributions to audio-visual speech recognition technologies. With a total of five patents to his name, Braga has made significant strides in enhancing the accuracy and privacy of speech recognition systems.

Latest Patents

Braga's latest patents include innovative methods that leverage audio and visual data to improve speech recognition. One notable patent is titled "Rescoring automatic speech recognition hypotheses using audio-visual matching." This method involves receiving audio data corresponding to an utterance spoken by a user and video data representing the motion of the user's lips while speaking. The process generates multiple candidate transcriptions for the utterance and determines an agreement score to select the most accurate transcription based on the lip movements.

Another significant patent is "Privacy-aware meeting room transcription from audio-visual stream." This method addresses privacy concerns by segmenting audio data and determining the identity of speakers based on image data. It ensures that privacy conditions requested by participants are respected during the transcription process.

Career Highlights

Otavio Braga is currently employed at Google Inc., where he continues to develop cutting-edge technologies in speech recognition. His work focuses on integrating audio and visual elements to create more effective communication tools.

Collaborations

Braga collaborates with talented individuals such as Takaki Makino and Richard Cameron Rose, contributing to a dynamic team that pushes the boundaries of speech recognition technology.

Conclusion

Otavio Braga's innovative work in audio-visual speech recognition has the potential to transform how we interact with technology. His patents reflect a commitment to enhancing both accuracy and privacy in communication systems.