Senior Scientist, Foundation models for speech

Specialista Senior / Project Manager About Translated Translated is on a mission to enable everyone to understand and be understood in their own language. We are a technology-driven professional translation provider partnering with over 200,000 translators worldwide in 200 languages. Our 310,000 clients range from individuals needing their CVs translated to major corporations like Uber and Airbnb. We leverage scientific progress and the synergy between humans and machines, investing heavily in R&D such as LLMs applied to translation, expressive speech synthesis, and privacy-preserving training. As a science-driven startup, we aim to quickly translate scientific innovations into impactful production tools. The project: Meetween Translated has received a grant for Meetween, a €7M, 4-year collaborative research project starting January 2024, which Translated leads. Meetween employs LLMs and multimodal foundation models to enhance human communication. The project focuses on Deep Learning, Large Language and Multimodal models, Machine Translation, Automatic Speech Recognition and Translation, Summarization, and AI Digital Assistants. It offers collaboration opportunities with leading academic and industry teams in speech processing. Our goal with Meetween is to "solve speech" by building foundation models that integrate text, audio, and video modalities (including lip movement, facial expressions, and gestures) into a single architecture. This enables downstream tasks like ASR, zero-shot TTS, voice cloning, speech-to-speech translation, lip reading, and audio/video enhancement. We have secured significant computing resources, including hundreds of thousands of A100 GPU-hours on Polish HPC infrastructure and in-house GPUs. All research outputs, including trained models, datasets, and benchmarks, will be open-sourced on HuggingFace. What You’ll Do You will join the AI Research team focused on Meetween, working on Large Language Models, Machine Translation, Speech Synthesis, and privacy-preserving Machine Learning. You will collaborate with product and engineering teams to develop next-generation technologies. Your responsibilities include: Working with data, compute, and algorithms Designing multimodal neural architectures Conducting experiments, coding, running large-scale evaluations Monitoring and benchmarking state-of-the-art research Guiding junior team members such as PhD students and interns Coordinating with partners on research roadmap Adapting to rapid scientific advances Organizing publications and open-source projects Requirements PhD or 4 years of industry research in relevant deep learning fields (e.g., language modeling, speech recognition) Strong programming skills in PyTorch Familiarity with Docker, Unix, GPU experiments Interest in experimental research Relevant publications, teaching, and research experience Experience in speech and language technology industry Team coordination skills Excellent English proficiency Expertise in multi-GPU training and optimization Polyglot abilities Open-source contributions Publications at top ML/AI conferences (NeurIPS, ICML, Interspeech, ACL, EMNLP) Translated is based at Pi Campus, a nature-immersed environment with villas in Rome, fostering talent and innovation. Pi Campus also reinvests in promising AI startups. Benefits include gym, swimming pool, kickboxing, water aerobics, fitness, Pilates, table tennis, football, kitchen, snacks, and bonuses for healthy habits and family growth. We celebrate diversity and are committed to creating an inclusive environment where everyone can thrive regardless of race, gender, or background. J-18808-Ljbffr

Location: Roma, Lazio, IT

Posted Date: 8/16/2025