This job might no longer be available.
Manager, Voice Input and Speech Synthesis
3 years ago
Job Description
In this position, you are responsible for owning and shipping Magic Leap’s embedded and cloud Voice Input and Speech Synthesis services. This includes end-to-end design, development and productization of multi-modal voice interactions, speech recognition (speech-to-text STT, ASR), natural language processing/understanding (NLP/NLU) and text-to-speech (TTS) services. As part of your job, you will be working with existing and pre-release Magic Leap devices on a daily basis.
Responsibilities
- Work with teams from the low-level audio sub-system, middleware services, SDK and cloud to define end-to-end architecture and dependency alignment for speech-to-text, ASR, NLU and TTS services
- Work with Product Management and UX designers to define and design features such as multi-modal voice interactions and the voice UI/UX
- Technical evaluation of TTS, ASR, NLU and STT solutions
- KPI definition for TTS, ASR, NLU and STT
- Product requirements analysis and conversion to architecture and software requirements
- Define and maintain development roadmap, resource and risk plans to meet product release milestones
- Approximately 30% of time devoted to hands on development and productization of voice input and speech synthesis services
- Build, grow, provide technical guidance and lead team of engineers responsible for
- Technical feasibility studies, proof of concepts and prototypes
- Design, implementation and productization of embedded (on-device) and cloud voice and speech platform services and API’s
- ASR/NLU model development, training and tuning
- Implementation of multi-modal input support
- Test automation for word error rate (WER), NLU intent accuracy, etc
Qualifications
- 7+ years of experience in software services development and productization, including 2+ years as technical/team lead in speech processing or related field
- Experience with and understanding of natural language understanding and/or speech recognition (ASR) systems and algorithms
- Proficient in Python/NodeJS, high-level familiarity with C/C++ is a strong plus
- Familiarity with middleware services development in embedded systems is a plus
Education
- BS or MS in Computer Science or related field
Additional Information
- All your information will be kept confidential according to Equal Employment Opportunities guidelines.
Create Your Profile — Game companies can contact you with their relevant job openings.