Audio AI Research Scientist — Audio Source Separation

Bose Corporation; Framingham (MA); On-site; USD 165,000 - 228,000; Full time

Job summary

Bose Corporation in Framingham, MA is seeking a Research Scientist in Audio Source Separation to develop real-time AI-powered audio algorithms with ML, DSP, and software colleagues. You will prototype, train, and evaluate novel source isolation and extraction methods for microphone recordings, integrating with embedded platforms and pursuing patents and conference papers.

We offer a competitive compensation range ($165,400–$227,500) and a comprehensive benefits package reflecting Bose’s

Qualifications

• Graduate degree (MS or PhD) in ML, EE, CS, or related field.
• Experience building audio source separation models and evaluating them.
• Strong knowledge of audio DSP and spatial signal processing.

Responsibilities

• Research, prototype, train, and evaluate source separation algorithms.
• Develop single- and multi-microphone approaches with spatial cues.
• Assess performance under noise, reverberation, and motion.
• Benchmark perceptual quality and latency for real-time use.
• Collaborate with DSP and embedded teams; contribute to patents and papers.

Skills

Audio source separation

Speech enhancement

DSP and spatial processing

PyTorch ML

C/C++ familiarity

Real-time inference

Education

MS or PhD in ML/EE/CS

Tools

ONNX/TFLite

Embedded deployment

Python

Job description

At Bose Corporation, we believe sound is the most powerful force on earth — and for over 60 years, we have been a company built on innovation, excellence, and independence. Privately owned, fiercely customer-focused, and driven by our values, we continue to lead industries and transform lives through sound.

Today, Bose Corporation is entering an exciting new era. Across multiple global Business Units and Global Functions, we are shaping the future of audio technology, automotive, luxury, and premium experiences. We invite you to join us in this transformation.

Job Description

Research Scientist — Audio Source Separation

Description

The goal of our team is to develop novel AI-powered audio processing algorithms. The twist is that our algorithms must run in real time, on physical devices, for applications such as dynamic awareness, hearing augmentation, agentic perception, voice pickup, content enhancement, and ones we haven’t even thought of yet. You will work with experts in ML, DSP, software engineering, and psychoacoustics to prototype and implement new algorithms.

You will develop next-generation algorithms that isolate, extract, or suppress individual sound sources from microphone recordings, sometimes augmented by other sensor signals. Applications include voice pickup, hearing augmentation, and new audio experiences that must ultimately run in real time on physical devices.

Bose has a strong history of combining creative thinking with cutting-edge audio technology. We are looking for people passionate about machine learning and audio to help shape what comes next.

Responsibilities

• Research, prototype, train, and evaluate novel source separation and extraction algorithms for speech and other real-world sounds.
• Develop approaches for single- and multi-microphone recordings, including spatial cues and optional auxiliary sensors.
• Investigate robust separation in challenging conditions such as overlapping talkers, noise, reverberation, and motion.
• Evaluate perceptual quality, separation performance, generalization, latency, memory, and computational requirements.
• Collaborate with DSP and embedded software teams to deploy models in real-time demonstrations and product platforms; contribute to patents and conference papers.

Education

Graduate degree (MS or PhD) in machine learning, electrical engineering, computer science, or a related field, or equivalent research experience.

Skills

• Deep hands-on experience in audio source separation, target sound extraction, speech enhancement, or related neural audio processing.
• Comfortable tackling complex, open-ended problems where you have to experiment with and discard many ideas before finding the right solution.
• Strong knowledge of audio DSP, time-frequency methods, and microphone array or spatial signal processing.
• Experience training and evaluating audio models in PyTorch; proficiency in Python and familiarity with C/C++ is a plus.
• Understanding of objective and perceptual evaluation methods and the tradeoffs involved in real-time low-power inference.
• Experience with multimodal or sensor-conditioned separation, ONNX/TFLite, or embedded deployment is a plus.
• Strong communication skills; publications at conferences such as ICASSP, Interspeech, NeurIPS, or similar are welcome but not required.

At Bose, you're inspired to be and do your best and are rewarded for your unique talents! Our compensation is thoughtfully tailored to your skills, experience, education, and location, and goes beyond base salary. The hiring range for this position in the primary work location of Framingham, Massachusetts is: $165,400-$227,500.The hiring range for other Bose work locations may vary. In addition to competitive base pay we offer rewards including bonus programs, comprehensive health and welfare benefits, a 401(k) plan, plus exclusive perks designed to support your wellbeing, and a generous employee discount where you can immerse yourself in our products and experiences. We are a proudly independent company—driven by purpose, guided by our values, and united by a belief in the power of sound. As the world leader in audio experiences, we’re creating what’s next—pushing boundaries and delivering transformative sound experiences for people everywhere. Join us and make your next career move a mic-drop. Let’s Make Waves.