If you are a current Motorola Solutions employee, please click this link to apply through your Workday account.Company Overview
At Motorola Solutions, we create technologies our customers refer to as their lifeline. Our technology platforms in communications, software, video and services help our customers work safely and more efficiently. Whether it’s helping firefighters see through smoke, enabling police officers to see around street corners, or reliably keeping the lights on in homes and businesses around the world, our work supports those who put their lives on the line to keep us safe. Bring your passion, potential and talents to Motorola Solutions, and help us usher in a new era in public safety and security.
Department OverviewMotorola Solutions’ innovations, products and services play essential roles in people’s lives. Our end-to-end suite of software solutions helps our customers answer thousands of emergency calls and text messages, and process video, evidence and records. We are also proud to be industry leaders in video security solutions installed in more than 120 countries at thousands of customer sites including school campuses, transportation systems, healthcare centers, public venues, critical infrastructure, prisons, factories, casinos, airports, financial institutions, government facilities, and retailers.
Our products leverage cutting edge voice technology, language models, natural language understanding, video analytics, and Machine Learning to enable our customers to focus on what matters while providing faster responses, safer outcomes and greater transparency. The candidate will join a talented team of scientists and engineers (located in Chicago, Boston, Vancouver, Krakow, and remote) in the Chief Technology Office at Motorola Solutions working on developing mission-critical AI solutions for our public safety and commercial customers. Our mission is to provide meaningful value to MSI’s customers by developing domain-specific applications and services in the areas of Speech & Audio, Machine Translation, Natural Language Understanding and Computer Vision.
Our AI team is focused on all aspects of 1) designing, prototyping and developing solutions (e.g. algorithms and architectures for object detection, classification etc.), 2) Machine Learning training and optimization (distributed training and inference acceleration) and 3) data (tools for data acquisition, annotation – both human and machine-powered, data pipelines and supporting ML infrastructure). We’re excited about the work we are doing in AI. We often describe it as “Mission Critical AI” because our approach is founded on our many years of Public Safety and Video Security experience and knowledge. We know the mission-critical needs of our customers are unique and different from the most consumer technologies that leverage AI today.
We are looking for a passionate and talented research scientist with a particular domain expertise in Speech & Audio technology. Our Speech & Audio team is responsible for quantitative data analysis, building, evaluating and deploying both semantic and non-semantic speech models for our public safety and commercial customers that scale to many thousands of hours of audio.
The role is focused in design and development of novel machine learning models to advance speech and audio processing technology that complements Motorola Solutions’ massive scale in mission critical communications. As a scientist on this team, you will work with talented peers to develop algorithms and modeling techniques and apply this effort to delivering experiences for the mission-critical and unique needs of our customers. You will be responsible for leading and collaborating on ambitious projects, including determining an appropriate scope for the project, defining relevant evaluation metrics based on the use case and user research, writing research-grade inference and training code, and thoroughly documenting and communicating your process and results to the team and management in order to push your projects through to a production environment. This role requires aptitude in experimental design, software engineering, machine learning, and communication of results.
More specifically, this involves:
Conducting research and development in cutting edge areas of the Speech & Audio domain, including ASR, sound classification, signal processing, audio characterization, and others that can be applied to MSI product development.
Understanding and implementing state-of-the-art algorithms for Speech & Audio applications that run on embedded devices, servers and cloud.
Defining custom use case appropriate evaluation metrics for algorithm performance based on statistical analysis of algorithm output.
Designing, prototyping, and adapting solutions in response to critical market needs defined by product management and senior management.
Writing research-grade code for deployment in research prototypes.
Working with scientists and engineers to minimize the footprint of trained ML models with various pruning, compression and real-time adaptation techniques.
Staying up-to-date with the latest research and competing technologies, including periodically attending conferences and reporting your findings back to the team and management.
Generating high quality patents and/or academic publications in the Speech & Audio domain as well as documenting the algorithms and software solutions developed as part of this process.
Excellent understanding of machine learning techniques and algorithms specifically applied to Speech & Audio tasks
Strong software engineering skills including python, unix command line tools and scripting, OOP, docker and/or kubernetes, git.
Strong mathematical background: knowledge of advanced calculus, linear algebra, probability theory, optimization and statistics
Experience with research-oriented implementation skills, such as fluency with libraries for scientific computing (e.g. scipy ecosystem) and machine learning (e.g. scikit-learn, PyTorch, TensorFlow)
Experience communicating research for public audiences of peers
Strong written and oral communication skills
Additional Desired Skills (Plus):
Hands-on experience in Speech & Audio related projects like ASR, speaker separation, diarization, signal processing, sound classification, MFCC/spectrogram feature representations, audio context understanding, ivector/xvector modeling, etc.
Experience with open source Speech & Audio frameworks like Kaldi.
Working knowledge of common python libraries (e.g. scikit-learn, numpy, scipy, matplotlib, pandas, jupyter etc.)
Experience with deep learning frameworks like pytorch, keras, tensorflow.
Activity in the Open Source community (e.g. github account as reference) is a strong plus.
Publications in leading workshops or conferences (e.g. NeurIPS, AAAI, WASPAA/SANE/ DCASE, etc).
Experience working with and analyzing large datasets in solving analytical problems using quantitative approaches.
- Graduate degree (Masters or PhD) in Machine Learning, Computer Science, Applied Mathematics or a relevant field
- 3+ years of experience with machine learning techniques and algorithms specifically applied to Speech & Audio tasks
Travel RequirementsUnder 10%
Referral Payment PlanYes
Motorola Solutions is an Equal Opportunity Employer and all qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, veteran's status, or, any other protected characteristic.