Research Staff, Voice AI Foundations

Position Overview

Deepgram is the leading voice AI platform for developers building speech-to-text (STT), text-to-speech (TTS) and full speech-to-speech (STS) applications. With over 200,000 developers and a track record of processing trillions of words, Deepgram is driving innovation in voice interaction. Operating with an AI-first mindset and a rapid pace of change, we are looking for a researcher to pioneer breakthrough solutions in voice AI by developing new Latent Space Models that overcome real-world audio challenges.

Key Responsibilities

• Build next-generation neural audio codecs that achieve extreme, low bit-rate compression and high-fidelity reconstruction across a world-scale corpus of general audio.
• Pioneer steerable generative models capable of synthesizing the full diversity of human speech—from casual conversation to emotionally charged expression and complex multi-speaker scenarios.
• Develop embedding systems to factorize codec latent spaces into interpretable dimensions such as speaker, content, style, environment, and channel effects, enabling precise control and data amplification through “latent recombination”.
• Leverage latent recombination to generate synthetic audio data at unprecedented scales and train multimodal speech-to-speech systems with empathic, human-like responses.
• Design model architectures, training schemes, and inference algorithms optimized for bare-metal hardware to enable cost-efficient training on billion-hour datasets and power real-time inference for massive concurrent conversations.

Required Qualifications

• Strong mathematical foundation in statistical learning theory, especially in areas pertinent to self-supervised and multimodal learning.
• Deep expertise in foundation model architectures with a proven track record of scaling training across multiple modalities.
• Proven ability to bridge theory and practice by deriving novel mathematical formulations and implementing them efficiently.
• Demonstrated experience in building data pipelines to process and curate massive, diverse datasets while ensuring data quality.
• Track record in designing controlled experiments to isolate the impact of architectural innovations and validate theoretical insights.
• Experience optimizing models for real-world deployment, including managing hardware constraints and efficiency techniques.
• History of open-source contributions or research publications that have advanced the state of the art in speech and language AI.

Benefits & Perks

• Holistic health: Medical, dental, vision benefits; annual wellness stipend; mental health support; life, STD, LTD Income Insurance Plans.
• Work/life blend: Unlimited PTO; generous paid parental leave; flexible schedule; 12 paid US company holidays; quarterly personal productivity stipend; one-time home office upgrade stipend; 401(k) plan with company match; tax savings programs.
• Continuous learning: Learning/education stipend; participation in talks and conferences; Employee Resource Groups; AI enablement workshops and sessions.