SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
DeepL is seeking a Senior Staff Research Scientist to lead scientific innovation across the DeepL Voice platform, which delivers real-time multilingual speech-to-speech translation. This is a hands-on leadership role combining deep technical expertise with strategic vision.
You will own research and development across the full speech translation stack: automatic speech recognition (ASR), machine translation (MT), text-to-speech (TTS), and end-to-end speech-to-speech systems. Key responsibilities include designing and training large-scale multilingual ASR models optimized for accuracy, robustness, and ultra-low-latency streaming; improving cascaded translation pipelines end-to-end (segmentation, ASR→MT interfaces, streaming inference, incremental decoding); developing real-time TTS models with natural prosody and fast inference; and building emerging end-to-end and LLM-based speech translation systems including streaming and one-shot approaches.
You will manage the complete model lifecycle from prototyping and ablation studies through training, evaluation, optimization, and production deployment. Close collaboration with engineering teams is essential to integrate models into reliable, scalable real-time systems. You'll drive improvements in inference efficiency, model serving, voice UX, and robustness to real-world acoustic conditions, while establishing strong practices for evaluation, reproducibility, monitoring, and continuous improvement in production.
Mentorship is a core part of this role—you'll guide researchers and engineers, promote hands-on collaboration, and elevate technical quality across the applied research team.
Required qualifications include deep expertise in speech, audio, or multilingual machine learning (ASR, MT, TTS, end-to-end speech translation, or large speech models); proven ability to train models, run experiments, debug pipelines, and integrate ML systems into production; strong understanding of real-time streaming constraints and low-latency model design; experience shipping and maintaining ML models at scale with engineers; ability to balance complex research with product impact and user experience; strong coding skills (Python, PyTorch/JAX, audio processing); clear communication and cross-team collaboration; and demonstrated mentorship experience.
