Remote job
Research Engineer, Speech
Job details
About this role
Role overview Develop speech and audio models for natural, continuous conversations, including interactions that adapt to pauses, interruptions, and changes in what a person is asking. The work spans research, data preparation, evaluation, and collaboration with inference and systems engineers to make models responsive on devices.
Responsibilities - Design and train models for speech generation, audio representations, or conversational behavior. - Build pipelines to clean, segment, align, and assess audio data. - Study how conversation history affects prosody, delivery, and consistency between turns. - Develop turn-taking, backchannel, and interruption handling, including recovery when requests change. - Evaluate intelligibility, voice quality, timing, and interaction failures through listening studies and objective measures. - Improve streaming generation with systems engineers, balancing responsiveness and output quality on target devices.
Requirements - Strong Python skills and practical experience with a modern deep-learning framework. - Expertise in at least one speech or audio area, such as synthesis, recognition, audio representations, or speech-to-speech systems. - Experience preparing audio data and designing experiments to diagnose model behavior. - Ability to connect model training with inference, evaluation, and an interactive prototype. - Careful listening and sound technical judgment about qualities that standard benchmarks may miss.
Nice to have - Experience with streaming architectures, neural audio codecs, waveform generation, or full-duplex systems. - Work involving multilingual speech, expressive synthesis, human preference studies, or low-latency deployment.
Benefits and work setup - Full-time role; based in San Francisco or remote.