We are looking for a Technical Program Manager in Speech and Audio System Quality to take charge in defining requirement for SONOS future line of products. The role also includes the collection, processing, and curation of speech and audio data with an emphasis on ultimate quality and user experience.
You will be using and developing state-of-the-art techniques to simulate, measure and predict how our products will perform in our customer’s homes. Your team will influence and shape our products, reduce our time to market and ensure that Sonos customers have a great overall experience.
* What You’ll Do
- Organize and execute data collection projects in varying acoustic environments and multiple languages.
- Coordinate end-to-end process of data preparation and design, execution, troubleshooting, postprocessing, and reporting.
- Work very closely with technical teams to define requirements, features, and annotation rules for data collection that develop cloud features.
- Hire and manage a team
- Be creative! Envision new processes and procedures for data collection to improve accuracy, and put those plans into action.
- Work with tools teams to improve the process and experience for data evaluators.
- Use a mix of manual, semiautomatic and fully automated test tools to gather quantitative results in the lab, through simulation and in the field.
- Collaborate with external voice service partners and vendors.
Qualifications required
- B.Sc. in CS/CE/EE or B.A. in Psychology/Linguistics. M.Sc. in an engineering discipline preferred.
- 5+ years managing and organizing large-scale speech and/or audio data collections.
- Managing vendors, and/or product development.
- 5+ years of Software Development, Software Operations, SQA.
- Good knowledge of C/C++, Python, Matlab/Simulink, preferably in Audio or Voice system development.
- Manage small teams and external contractors.
- Manual testing and small to large scale test automation.
- Handle large amounts of quantitative data and aggregate results that are meaningful to developers.
- Be very analytical, data driven, self-motivated and pro-active, with multi-tasking and creative and critical thinking capabilities.
* Experience in one or more of these areas is a definite plus:
- Speech and audio quality metrics (PESQ, POLQA, ViSQOL, etc.),
- audio and speech processing algorithms and machine learning technology.
- speech recognition toolkits (e.g., KALDI, HTK) or other
- be very creative, find new innovate ways to test and measure voice performance
Experience with the following tools: ACQUA (HEAD acoustics), Audio Precision, HATS, and Audition.