Embedded Audio DSP Engineer

San Francisco, CA, United States


Role Location

  • San Francisco, CA, United States


101 - 250 people


400 Treat Ave Ste F
San Francisco, CA, 94110, US

Tech Stack

  • C
  • C++
  • Python
  • Tensorflow

Role Description


As the Audio DSP Engineer at Whisper, you’ll be responsible for developing new algorithms for a next generation wireless hearing aid platform. This will include developing novel techniques in acoustic modeling as well as tweaking existing algorithms to be best optimized for our hardware platform and consumer use case.

This role will span both algorithms research and prototyping (e.g. MATLAB or Python) as well as writing shipping code on the target hardware (e.g. Assembly and C) to bring these algorithms to customers.


Develop new techniques in acoustic modeling optimized for a hearing aid system that includes deep learning audio processing. Optimize existing (either in-house or vendor-provided) algorithms to our hearing aid system. Debug and improve audio processing performance on target hardware using manufacturing samples. Receive feedback from customers improve the hearing experience in the Whisper hearing aid system in conjunction with the machine learning acoustics team.


Product focused: Excited to work at a young company and take a major role in developing a new hardware product that will help people hear better in everyday situations. 3+ years of industry experience in designing and implementing audio DSP algorithms related to audio and/or speech processing in a hardware product. Strong theoretical background of signal processing techniques including adaptive filtering, filter banks and wavelet processing, speech analysis and synthesis, speech and audio coding. Comfortable prototyping audio algorithms (e.g. MATLAB or Python) and bringing these algorithms to the DSP by writing customer-facing code (e.g. Assembly and C). Understanding of analog and digital audio processing (in particular filter theory and design) and adaptive audio processing algorithms. Experience working in a team software engineering environment (e.g. code reviewing, source control, etc). BONUS POINTS

Familiar with traditional statistical modeling/feature extraction techniques (GMM, HMM, NMF / spectrograms, MFCC etc) for voice recognition and audio event classification. Experience with machine learning techniques (e.g. deep neural nets) to solve audio tasks like audio event classification or voice detection. Electrical (analog and digital) audio design knowledge. Audio qualification experience, including calibration procedures, test procedures, equipment, tools, etc.

About Whisper

Whisper exists to improve the human senses. Our hearing aids uses deep learning to remove background noise so you can hear clearly in a restaurant or at home when the dishwasher is running – this is the number one complaint in the industry, and our recent results reduce unwanted noise by 10x any other brand.

Interested in this role?
Skip straight to final-round interviews by applying through Triplebyte.