LingCircle Roundtable: Insights on Linguistics Roles in Speech Recognition Technology
The SoundHound team in Boulder has been growing and is excited to share updates since last year’s LingCircle! There will be a brief overview of what SoundHound does and how the linguistic roles fit in. It will then open up to questions and discussion with the team, followed by a social hour with light bites. All in the new SoundHound Boulder office (see address below)! This is also an opportunity to explore possible linguistic summer internships in the SoundHound Boulder office! Attendance is limited, so sign up now at if interested in attending.Â
Friday, January 31, 2020
4:00-5:30 PM Presentation and roundtable discussion
5:30-6:30 PM Social hour with light bites
SoundHound Boulder
1468 Pearl St, Suite 200
Abstract:
At SoundHound, we strive for success at every level, starting with a critical focus on finding the best possible match in the folks we hire. This can be challenging when it comes to college candidates with very little or no industry experience, especially because the speech recognition industry is still in its infancy and not very well defined. Arming linguistic college candidates with information about the skills and experience we look for, as well as details about the work we do, helps ensure success for both SoundHound and for our college linguist hires. Â
All questions welcome! Here are some examples from last year:
1.      What do you look for on a resume and during interviews?
2.      How and why did you choose your specific degree and career path?
3.      What are the qualifications needed to work in your field?
4.      What is a typical work day like for you?
5.      What do you find most challenging and most rewarding about your work?
6.      What do you wish you would have known before entering your career?
SoundHound Biographies:
Lisa Rutta
Lisa is the Director, Ops and Technical Project Management at SoundHound. She leads the Language Data Programs team at SoundHound which she has grown from the ground up. The team owns the data used to train SoundHound's machine learning speech system, including the curation, validation, annotation and transcription of text and recorded audio data. Her team includes language evaluators, degreed linguists and SQL/Unix/Python coders, and delivers projects that directly impact the performance, scale and the quality of our speech recognition and music discovery systems! Lisa has a BS in Computer Science and Human Computer Interaction. She shifted focus early in her career from software development to product and project management. She has experience growing teams and processes at large companies with both consumer and business-business customers (Xerox, Intuit and Yahoo), and also at startups, such as SoundHound.Â
Jenette Preciado
Jenette Preciado is a project coordinator for the Language Data Programs team, focusing primarily on improving the quality of data annotation for English-specific projects. In a typical week, she meets with a project’s stakeholders to discuss data requirements, resources, and timelines in an effort to design and apply effective action plans. She is a University of Colorado Boulder alum (Go Buffs!) and Ronald E. McNair scholar with a bachelor’s degree in linguistics, art history, and Spanish language and literature.Â
Kiersten Bradley
Kiersten Bradley is a project coordinator for the Language Data Programs team. She focuses on the management of annotators, communicating wants, needs, and expectations between clients, SoundHound, and data annotators in order to create processes benefiting both current milestones and future expansion. Her favorite aspect of her position is SoundHound’s holistic approach to its employees, encouraging them to learn new technical skills, collaborate with other projects, and creatively assess and solve problems. She holds bachelor’s degrees in English and Anthropology from the University of Florida and the University of North Florida, respectively, as well as a master’s degree in Linguistics and a Culture, Language, and Social Practice (CLASP) certificate from the University of Colorado Boulder.Â
Anton
Anton Relin is a member of the Hound team at SoundHound and is involved in creating domain specific natural language understanding solutions. Day to day work includes creating graphs to model English language queries, and software engineering to produce meaningful responses based upon the utterances. He received his Bachelor’s degrees in Linguistics, Computer Science, and Slavic Studies from the University of Pennsylvania, where he concentrated on the intersection of linguistics and computer science, and on the Russian avant-garde of the 1910s.Â
Maisy Wieman
Maisy Wieman is a part of the Speech EngineeringTeam at SoundHound, where she uses machine learning and signal processing to model phonemes across different languages. She received her Bachelor's and Master's degree in Electrical Engineering from Stanford University. Her concentration was in signal processing, which she applies to her work in the context of speech enhancement. In addition to addressing recognition of noisy and reverberant audio, she designs and trains neural networks to perform accurate phoneme classification. The one thing she loves even more than the Fourier Transform is a good book, so when she's not tackling a new engineering challenge, she's usually curled up with a novel.Â
About SoundHound:
SoundHound Inc. turns sound into understanding and actionable meaning. We believe in enabling humans to interact with the things around them in the same way we interact with each other: by speaking naturally to mobile phones, cars, TVs, music speakers, and every other part of the emerging ‘connected’ world. Our consumer product, Hound, leverages our Speech-to-Meaning™ and Deep Meaning Understanding™ technologies to create a groundbreaking smartphone experience, and is the first product to build on the Houndify platform. Our SoundHound product applies our technology to music, enabling people to discover, explore, and share the music around them, and even find the name of that song stuck in their heads by singing or humming. Through the Houndify platform and Collective AI, we aim to bring voice-enabled AI to everyone and enable others to build on top of it. Our mission: Houndify everything.