This inventor holds 1 USPTO granted patent. Top assignee: Amazon Technologies, Inc.. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Bartosz Putrycz and Robust Speech Synthesis
Introduction
Bartosz Putrycz is an inventor based in Gdansk, Poland, associated with Amazon Technologies, Inc. His patent work addresses the challenge of making speech synthesis more reliable when input text contains spelling, punctuation, or formatting variations.
Latest Patents
Putrycz is associated with 1 patent, titled “Speech Synthesis with Robustness Against Input Variation.” The invention describes a speech synthesis system designed to handle variations and errors in written input. A text modifier creates a parallel training dataset by introducing changes to text, and the system generates speech from the modified examples. A robustness tester compares that synthesized audio with audio from the original training data. The results can guide updates to speech-generation model parameters and help adjust how often text modifications are introduced, reducing the risk of harming performance on unmodified text.
Career Highlights
Putrycz’s listed patent reflects work on improving the resilience of speech synthesis systems. His affiliation with Amazon Technologies, Inc. connects this invention to the company’s technology development activities.
Collaborations
The listed coworkers associated with Putrycz include Yang Li and Mateusz Aleksander Lajszczak. Their names provide context for the collaborative environment surrounding the patent, though the supplied information does not specify their individual contributions.
Conclusion
Bartosz Putrycz’s patent focuses on helping speech synthesis systems produce reliable results despite variations in spelling, punctuation, and formatting.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
