This internship can be based in either Cambridge, UK or Aachen, Germany \\n\\nThe Machine Translation and Multilingual AI team is looking for research interns to explore open problems in speech-to-speech translation: how to build models that let people talk naturally with anyone, in any language, in real time.
Speech-to-speech translation is moving from cascaded pipelines to end-to-end models that listen, translate and speak in a single system. Many core questions remain open. How do you translate simultaneously, before the speaker has finished? How do you keep translated speech natural and expressive? How do large multimodal models best combine speech understanding, translation and speech generation? And how should these systems be evaluated?\\n\\nAs an intern, you will own a focused research project on one of these questions, working closely with researchers and engineers on the team. You will have access to large-scale data and compute, and we encourage publishing results at top venues.\\n\\nExample research directions include:\\n\\n- Streaming and simultaneous speech translation: policies for when to listen and when to speak, and the trade-off between quality and latency.\\n- Speech generation for translation: natural, expressive and prosodically appropriate output, including speech tokenizers and audio codecs.\\n- Multimodal foundation models for speech: how pre-training, fine-tuning and reinforcement learning shape speech-to-speech ability across languages.\\n- Multilingual scaling: extending to new languages and handling code-switching and unknown source languages.\\n- Evaluation: automatic metrics and human evaluation methods that reflect real conversational experience.
Define a research question with your mentor, review related work, and propose an approach.\\nDesign and run experiments on large-scale speech and translation models.\\nAnalyse results rigorously and share findings with the team.\\nWrite up your work, with the goal of a publication or a reusable research contribution.
Currently enrolled in a PhD or Master"s program in Computer Science, Electrical Engineering, Linguistics or a related field.\\nResearch experience in at least one of: speech recognition, speech synthesis, machine translation, NLP or multimodal learning.\\nHands-on experience with deep learning toolkits such as PyTorch, JAX or TensorFlow.\\nStrong programming skills in Python.
Research experience in speech-to-speech or speech-to-text translation, especially end-to-end, streaming or simultaneous approaches.\\nResearch experience in text-to-speech or speech generation, such as neural TTS, speech tokenizers, audio codecs, or expressive and multilingual synthesis.\\nExperience training or fine-tuning large language models or multimodal (audio + text) models.\\nFamiliarity with speech and translation evaluation, such as WER, BLEU/chrF/COMET-style metrics, MOS and human preference studies, and latency metrics.\\nPublications at relevant venues (e.g. Interspeech, ICASSP, ACL, EMNLP, NeurIPS, ICML, ICLR).\\nExcellent spoken and written communication skills.
Speech-to-speech translation is moving from cascaded pipelines to end-to-end models that listen, translate and speak in a single system. Many core questions remain open. How do you translate simultaneously, before the speaker has finished? How do you keep translated speech natural and expressive? How do large multimodal models best combine speech understanding, translation and speech generation? And how should these systems be evaluated?\\n\\nAs an intern, you will own a focused research project on one of these questions, working closely with researchers and engineers on the team. You will have access to large-scale data and compute, and we encourage publishing results at top venues.\\n\\nExample research directions include:\\n\\n- Streaming and simultaneous speech translation: policies for when to listen and when to speak, and the trade-off between quality and latency.\\n- Speech generation for translation: natural, expressive and prosodically appropriate output, including speech tokenizers and audio codecs.\\n- Multimodal foundation models for speech: how pre-training, fine-tuning and reinforcement learning shape speech-to-speech ability across languages.\\n- Multilingual scaling: extending to new languages and handling code-switching and unknown source languages.\\n- Evaluation: automatic metrics and human evaluation methods that reflect real conversational experience.
Define a research question with your mentor, review related work, and propose an approach.\\nDesign and run experiments on large-scale speech and translation models.\\nAnalyse results rigorously and share findings with the team.\\nWrite up your work, with the goal of a publication or a reusable research contribution.
Currently enrolled in a PhD or Master"s program in Computer Science, Electrical Engineering, Linguistics or a related field.\\nResearch experience in at least one of: speech recognition, speech synthesis, machine translation, NLP or multimodal learning.\\nHands-on experience with deep learning toolkits such as PyTorch, JAX or TensorFlow.\\nStrong programming skills in Python.
Research experience in speech-to-speech or speech-to-text translation, especially end-to-end, streaming or simultaneous approaches.\\nResearch experience in text-to-speech or speech generation, such as neural TTS, speech tokenizers, audio codecs, or expressive and multilingual synthesis.\\nExperience training or fine-tuning large language models or multimodal (audio + text) models.\\nFamiliarity with speech and translation evaluation, such as WER, BLEU/chrF/COMET-style metrics, MOS and human preference studies, and latency metrics.\\nPublications at relevant venues (e.g. Interspeech, ICASSP, ACL, EMNLP, NeurIPS, ICML, ICLR).\\nExcellent spoken and written communication skills.