Principal Speech Scientist

Description

*Full-time with benefits

What you will do

You will lead the development of advanced models that derive meaningful insights from human speech. Your work will span acoustic feature extraction, voice biomarker analysis, paralinguistic understanding, and predictive modeling using state-of-the-art machine learning and DSP techniques. This is a hands-on, research-driven role where you will push the boundaries of voice-based AI in close collaboration with engineering, product, and research teams.

Responsibilities

• Design, train, evaluate, and deploy machine learning models for speech and voice analytics, including acoustic modeling, speaker embeddings, prosody/emotion analysis, and anomaly detection.
• Build DSP pipelines for feature extraction, noise reduction, signal enhancement, and preprocessing.
• Conduct large-scale experiments, define hypotheses, build and curate datasets, and perform statistical analyses to validate model performance and robustness.
• Explore, implement, and optimize state-of-the-art architectures such as self-supervised audio models, transformer encoders, and probabilistic or generative approaches.
• Collaborate with engineering teams to transition research into scalable, low-latency inference systems for production.
• Measure and optimize performance across accuracy, fairness/bias mitigation, computational efficiency, latency, and interpretability.
• Produce high-quality research artifacts, such as whitepapers, internal reports, patents, or scientific summaries.
• Stay informed on emerging innovations in speech AI, voice biometrics, generative audio, multilingual modeling, and related fields.
• Perform additional duties as required.

Who you are

We value team members who are innovative, driven, and enthusiastic. You are a curious, experimental, and scientifically rigorous thinker who thrives at the intersection of research and real-world impact. You’re comfortable owning projects end-to-end—from early exploratory analysis to building production-ready models—and you’re passionate about solving complex problems where the human voice is the core data source. You collaborate effectively across disciplines, communicate insights clearly, and excel in fast-paced environments where innovation, iteration, and teamwork drive meaningful outcomes.

Requirements

• Bachelor's in Computer Science, Machine Learning, Signal Processing, Computational Linguistics, Applied Math, or a related field or equivalent experience. MSc or PhD preferred.
• At least 14 years of experience with speech or voice ML: ASR, acoustic modeling, speaker modeling, prosody, paralinguistic features, or related domains.
• Strong foundation in DSP principles such as filtering, sampling, and acoustic feature extraction (e.g., MFCCs, spectral features).
• Proven experience building and deploying ML models using Python and modern ML frameworks (PyTorch, TensorFlow, JAX, etc.).
• Experience working with audio datasets, annotation pipelines, and large-scale experimental workflows.
• Strong statistical reasoning, including hypothesis testing and experimental design.
• Ability to clearly communicate scientific results to both technical and non-technical audiences.
• Experience with self-supervised or foundation audio models (e.g., wav2vec, HuBERT, Whisper, AudioLM) is preferred.
• Knowledge of telephony codecs, low-bitrate audio modeling, noise-robust processing, or real-time inference is preferred.
• Familiarity with edge/on-device deployment for speech or audio models is preferred.
• Experience with cloud infrastructure, GPU computing, and MLOps pipelines is preferred.
• Publications, patents, or open-source contributions in speech or voice AI is preferred.
• Experience working with multilingual, cross-cultural, or domain-specific speech datasets is preferred.
• Due to the nature of our US Government contracts, applicants must meet eligibility requirements for US Government security clearance. *Must be a US citizen and eligible for a US Government security clearance.*

Who we are

At Clearspeed, we are driven by corporate DNA committed to the service of others and a passion for our AI-enabled technology that is redefining risk assessment. Our fast-growing team spans operations across the US, Canada, and EMEA. Together our team of brilliant minds, diverse in expertise and experience, collaborate and contribute to a shared vision of enabling trust, faster.

Our company is committed to providing equal employment opportunities to all individuals. We value diversity and inclusion in our workplace, and we welcome and encourage individuals of all backgrounds to apply and join us on our journey to build trust faster.

Why join us?

*Impactful work

*Collaborative environment

*Work-life balance

*Remote/hybrid work flexibility

Our benefits (may vary based on geographical location)

*Competitive compensation: salary + performance-based bonuses

*Stock options

*Unlimited paid time off

*Health and wellness coverage

Join us at Clearspeed and be a part of our success story. Together, we can make a difference!

Salary

Salary range is based on national benchmark data of comparable roles, skills and experience level. Exact compensation will be based on individual skills and experience, and location which will be assessed during the interview process.

Salary Description

Salary range: $213,000 - $280,000