Social
Conversational AI

A quick guide to Amazon’s 40-plus papers at Interspeech 2022

Speech recognition and text-to-speech predominate, but other topics include audio watermarking, automatic dubbing, and compression.

By Staff writer
September 6, 2022
6 min read
Share

Of Amazon’s more than 40 papers at this year’s Interspeech, automatic speech recognition and text-to-speech account for about half. But the others cover a range of topics, from acoustic watermarking and automatic dubbing to quantization and fairness.

Acoustic watermarking

Practical over-the-air perceptual acoustic watermarking
Ameya Agaskar

Audio classification

CNN-based audio event recognition for automated violence classification and rating for Prime Video content
Tarun Gupta, Mayank Sharma, Kenny Qiu, Xiang Hao, Raffay Hamid

Impact of acoustic event tagging on scene classification in a multi-task learning framework
Rahil Parikh, Harshavardhan Sundar, Ming Sun, Chao Wang, Spyros Matsoukas

Automatic dubbing

Isochrony-aware neural machine translation for automatic dubbing
Derek Tam, Surafel Melaku Lakew, Yogesh Virkar, Prashant Mathur, Marcello Federico

Prosodic alignment for off-screen automatic dubbing
Yogesh Virkar, Marcello Federico, Robert Enyedi, Roberto Barra-Chicote

Automatic speech recognition

Compute cost amortized transformer for streaming ASR
Yi Xie, Jonathan Macoskey, Martin Radfar, Feng-Ju Chang, Brian King, Ariya Rastrow, Athanasios Mouchtaris, Grant Strimel

Cost amortized Transformer.png
"Compute cost amortized Transformer for streaming ASR" proposes a mechanism that toggles components of Transformer blocks on and off to use computational resources more efficiently.

Content-context factorized representations for automated speech recognition
David M. Chan, Shalini Ghosh

ConvRNN-T: Convolutional augmented recurrent neural network transducers for streaming speech recognition
Martin Radfar, Rohit Barnwal, Rupak Vignesh Swaminathan, Feng-Ju Chang, Grant Strimel, Nathan Susanj, Athanasios Mouchtaris

Directed speech separation for automatic speech recognition of long-form conversational speech
Rohit Paturi, Sundararajan Srinivasan, Katrin Kirchhoff, Daniel Garcia-Romero

Domain prompts: Towards memory and compute efficient domain adaptation of ASR systems
Saket Dingliwa, Ashish Shenoy, Sravan Bodapati, Ankur Gandhe, Ravi Teja Gadde, Katrin Kirchhoff

Incremental learning for RNN-Transducer based speech recognition models
Deepak Baby, Pasquale D'Alterio, Valentin Mendelev

Knowledge distillation via module replacing for automatic speech recognition with recurrent neural network transducer
Kaiqi Zhao, Hieu Duy Nguyen, Animesh Jain, Nathan Susanj, Athanasios Mouchtaris, Lokesh Gupta, Ming Zhao

Learning to rank with BERT-based confidence models in ASR rescoring
Ting-Wei Wu, I-FAN CHEN, Ankur Gandhe

Reducing geographic disparities in automatic speech recognition via elastic weight consolidation
Viet Anh Trinh, Pegah Ghahremani, Brian King, Jasha Droppo, Andreas Stolcke, Roland Maas

RefTextLAS: Reference text biased listen, attend, and spell model for accurate reading evaluation
Phani Sankar Nidadavolu, Na Xu, Nick Jutila, Ravi Teja Gadde, Aswarth Abhilash Dara, Joseph Savold, Sapan Patel, Aaron Hoff, Veerdhawal Pande, Kevin Crews, Ankur Gandhe, Ariya Rastrow, Roland Maas

RNN-T lattice enhancement by grafting of pruned paths
Mirek Novak, Pavlos Papadopoulos

Using data augmentation and consistency regularization to improve semi-supervised speech recognition
Ashtosh Sapru

Dialogue

Contextual acoustic barge in classification for spoken dialog systems
Dhanush Bekal, Sundararajan Srinivasan, Sravan Bodapati, Srikanth Ronanki, Katrin Kirchhoff

Adversarial reweighting.png
The method presented in "Adversarial reweighting for speaker verification fairness" uses an adversarial network to identify underperforming groups in a speaker verification dataset (green) and adjusts their contribution to the training loss (bottom).

Fairness

Toward fairness in speech recognition: Discovery and mitigation of performance disparities
Pranav Dheram, Murugesan Ramakrishnan, Anirudh Raju, I-Fan Chen, Brian King, Katherine Powell, Melissa Saboowala, Karan Shetty, Andreas Stolcke

Keyword spotting

Latency control for keyword spotting
Christin Jose, Joe Wang, Grant Strimel, Mohammad Omar Khursheed, Yuriy Mishchenko, Brian Kulis

Language identification

A multimodal strategy for singing language identification
Wo Jae Lee, Emanuele Coviello

Multidevice processing

Challenges and opportunities in multi-device speech processing
Gregory Ciccarelli, Jarred Barber, Arun Nair, Israel Cohen, Tao Zhang

Multiparty speech

Separator-transducer-segmenter: Streaming recognition and segmentation of multi-party speech
Ilya Sklyar, Anna Piunova, Christian Osendorfer

Natural-language understanding

Phonetic embedding for ASR robustness in entity resolution
Xiaozhou Zhou, Ruying Bao, William M. Campbell

Quantization

Squashed weight distribution for low bit quantization of deep models
Nikko Ström, Haidar Khan, Wael Hamza

Sub-8-bit quantization aware training for 8-bit neural network accelerator with on device speech recognition
Kai Zhen, Hieu Duy Nguyen, Raviteja Chinta, Nathan Susanj, Athanasios Mouchtaris, Tariq Afzal, Ariya Rastrow

Sub-8-bit quantization.png
The training behavior of the algorithm proposed in "Sub-8-bit quantization aware training for 8-bit neural network accelerator with on device speech recognition", in which weights are optimized to lower quantization loss.

Signal processing

Clock skew robust acoustic echo cancellation
Karim Helwani, Erfan Soltanmohammadi, Michael M. Goodwin, Arvindh Krishnaswamy

Real-time packet loss concealment with mixed generative and predictive model
Jean-Marc Valin, Ahmed Mustafa, Christopher Montgomery, Timothy B. Terriberry, Michael Klingbeil, Paris Smaragdis, Arvindh Krishnaswamy

Speaker identification/verification

Adversarial reweighting for speaker verification fairness
Minho Jin, Chelsea J.-T. Ju, Zeya Chen, Yi Chieh Liu, Jasha Droppo, Andreas Stolcke

Graph-based multi-view fusion and local adaptation: Mitigating within household confusability for speaker identification
Long Chen, Yixiong Meng, Venkatesh Ravichandran, Andreas Stolcke

Graph fusion and fairness.png
The method proposed in "Graph-based multi-view fusion and local adaptation" propagates labels across a graph whose nodes represent utterances and whose weighted edges quantify the similarity between utterances.

Spoken-language understanding

Learning under label noise for robust spoken language understanding systems
Anoop Kumar, Pankaj Sharma, Aravind Illa, Sriram Venkatapathy, Subhrangshu Nandi, Pritam Varma, Anurag Dwarakanath, Aram Galstyan

On joint training with interfaces for spoken language understanding
Anirudh Raju, Milind Rao, Gautam Tiwari, Pranav Dheram, Bryan Anderson, Zhe Zhang, Chul Lee, Bach Bui, Ariya Rastrow

Text-to-speech

Automatic evaluation of speaker similarity
Kamil Deja, Ariadna Sanchez, Julian Roth, Marius Cotescu

CopyCat2: A single model for multi-speaker TTS and many-to-many fine-grained prosody transfer
Sri Karlapati, Penny Karanasou, Mateusz Lajszczak, Ammar Abbas, Alexis Moinet, Peter Makarov, Ray Li, Arent van Korlaar, Simon Slangen, Thomas Drugman

Creating new voices.png
Voices created through the method presented in "Creating new voices using normalizing flows" (green) are spread across the embedding space of voices from the training set (blue), confirming that the method can generate a variety of new voices.

Creating new voices using normalizing flows
Piotr Biliński, Tom Merritt, Abdelhamid Ezzerg, Kamil Pokora, Sebastian Cygert, Kayoko Yanagisawa, Roberto Barra-Chicote, Daniel Korzekwa

Cross-lingual style transfer with conditional prior VAE and style loss
Dino Ratcliffe, You Wang, Alex Mansbridge, Penny Karanasou, Alexis Moinet, Marius Cotescu

End-to-end LPCNet: A neural vocoder with fully-differentiable LPC estimation
Krishna Subramani, Jean-Marc Valin, Umut Isik, Paris Smaragdis, Arvindh Krishnaswamy

Expressive, variable, and controllable duration modelling in TTS
Ammar Abbas, Tom Merritt, Alexis Moinet, Sri Karlapati, Ewa Muszynska, Simon Slangen, Elia Gatti, Thomas Drugman

GlowVC: Mel-spectrogram space disentangling model for language-independent text-free voice conversion
Magdalena Proszewska, Grzegorz Beringer, Daniel Saez Trigueros, Tom Merritt, Abdelhamid Ezzerg, Roberto Barra-Chicote

L2-GEN: A neural phoneme paraphrasing approach to L2 speech synthesis for mispronunciation diagnosis
Daniel Zhang, Ashwinkumar Ganesan, Sarah Campbell, Daniel Korzekwa

Low data? No problem: low resource, language-agnostic conversational text-to-speech via F0- conditioned data augmentation
Giulia Comini, Goeric Huybrechts, Manuel Sam Ribeiro, Adam Gabrys, Jaime Lorenzo Trueba

Mix and match: An empirical study on training corpus composition for polyglot text-to-speech (TTS)
Ziyao Zhang, Alessio Falai, Ariadna Sanchez, Orazio Angelini, Kayoko Yanagisawa

Simple and effective multi-sentence TTS with expressive and coherent prosody
Peter Makarov, Ammar Abbas, Mateusz Lajszczak, Arnaud Joly, Sri Karlapati, Alexis Moinet, Thomas Drugman, Penny Karanasou

Unify and conquer: How phonetic feature representation affects polyglot text-to-speech (TTS)
Ariadna Sanchez, Alessio Falai, Ziyao Zhang, Orazio Angelini, Kayoko Yanagisawa

Research areas
Tags
About the Author
Staff writer

Related content

Work with us

See more jobs See more jobs
Applied Scientist, Skill Builder, AWS Training and Certifications
US, VA, Arlington
Are you fascinated by the power of Large Language Models (LLM) and Artificial Intelligence (AI) to transform the way we learn and interact with technology? Are you passionate about applying advanced machine learning (ML) techniques to solve complex challenges in the cloud learning space? If so, AWS Training & Certification (T&C) team has an exciting opportunity for you as an Applied Scientist. At AWS T&C, we strive to be leaders in not only how we learn about the latest AI/ML development and AWS services, but also how the same technologies transform the way we learn about them. As an Applied Scientist, you will join a talented and collaborative team that is dedicated to driving innovation and delivering exceptional experiences in our Skill Builder platform for both new learners and seasoned developers. You will be a part of a global team that is focused on transforming how people learn. The position will interact with global leaders and teams across the globe as well as different business and technical organizations. Join us at the AWS T&C Science Team and become a part of a global team that is redefining the future of cloud learning. With access to vast amounts of data, exciting new technology, and a diverse community of talented individuals, you will have the opportunity to make a meaningful impact on the ways how worldwide learners engage with our learning system and builders develop on our platform. Together, we will drive innovation, solve complex problems, and shape the future of future-generation cloud builders. Please visit https://skillbuilder.awsto learn more. Key job responsibilities - Apply your expertise in LLM to design, develop, and implement scalable machine learning solutions that address challenges in discovery and engagement for our international audiences. - Collaborate with cross-functional teams, including software engineers, data engineers, scientists, and product managers, to define project requirements, establish success metrics, and deliver high-quality solutions. - Conduct thorough data analysis to gain insights, identify patterns, and drive actionable recommendations that enhance operational performance and customer experiences across Skill Builder. - Continuously explore and evaluate state-of-the-art techniques and methodologies to improve the accuracy and efficiency of AI/ML systems. - Communicate complex technical concepts effectively to both technical and non-technical stakeholders, providing clear explanations and guidance on proposed solutions and their potential impact. About the team Why AWS? Amazon Web Services (AWS) is the world’s most comprehensive and broadly adopted cloud platform. We pioneered cloud computing and never stopped innovating — that’s why customers from the most successful startups to Global 500 companies trust our robust suite of products and services to power their businesses. Inclusive Team Culture Here at AWS, it’s in our nature to learn and be curious. Our employee-led affinity groups foster a culture of inclusion that empower us to be proud of our differences. Ongoing events and learning experiences, including our Conversations on Race and Ethnicity (CORE) and AmazeCon conferences, inspire us to never stop embracing our uniqueness. Diverse Experiences AWS values diverse experiences. Even if you do not meet all of the qualifications and skills listed in the job description, we encourage candidates to apply. If your career is just starting, hasn’t followed a traditional path, or includes alternative experiences, don’t let it stop you from applying. Mentorship & Career Growth We’re continuously raising our performance bar as we strive to become Earth’s Best Employer. That’s why you’ll find endless knowledge-sharing, mentorship and other career-advancing resources here to help you develop into a better-rounded professional. Work/Life Balance We value work-life harmony. Achieving success at work should never come at the expense of sacrifices at home, which is why we strive for flexibility as part of our working culture. When we feel supported in the workplace and at home, there’s nothing we can’t achieve in the cloud.
Research Scientist, Intelligent Talent Acquisition, Global Hiring Science
US, VA, Arlington
Do you want a role with deep meaning and the ability to have a global impact? Hiring top talent is not only critical to Amazon’s success – it can literally change the world. It took a lot of great hires to deliver innovations like AWS, Prime, and Alexa, which make life better for millions of customers around the world. As part of the Intelligent Talent Acquisition (ITA) team, you'll have the opportunity to reinvent Amazon’s hiring process with unprecedented scale, sophistication, and accuracy. ITA is an industry-leading people science and technology organization made up of scientists, engineers, analysts, product professionals, and more. Our shared goal is to fairly and precisely connect the right people to the right jobs. Last year, we delivered over 6 million online candidate assessments, driving a merit-based hiring approach that gives candidates the opportunity to showcase their true skills. Each year we also help Amazon deliver billions of packages around the world by making it possible to hire hundreds of thousands of associates in the right quantity, at the right location, at exactly the right time. You’ll work on state-of-the-art research with advanced software tools, new AI systems, and machine learning algorithms to solve complex hiring challenges. Join ITA in using cutting-edge technologies to transform the hiring landscape and make a meaningful difference in people's lives. Together, we can solve the world's toughest hiring problems. Within ITA, the Global Hiring Science (GHS) team designs and implements innovative hiring solutions at scale. We work in a fast-paced, global environment where we use research to solve complex problems and build scalable hiring products that deliver measurable impact to our customers. We are seeking selection researchers with a strong foundation in hiring assessment development, legally-defensible validation approaches, research and experimental design, and data analysis. Preferred candidates will have experience across the full hiring assessment lifecycle, from solution design to content development and validation to impact analysis. We are looking for equal parts researcher and consultant, who is able to influence customers with insights derived from science and data. You will work closely with cross-functional teams to design new hiring solutions and experiment with measurement methods intended to precisely define exactly what job success looks like and how best to predict it. Key job responsibilities What you’ll do as a GHS Research Scientist: • Design large-scale personnel selection research that shapes Amazon’s global talent assessment practices across a variety of topics (e.g., assessment validation, measuring post-hire impact) • Partner with key stakeholders to create innovative solutions that blend scientific rigor with real-world business impact while navigating complex legal and professional standards • Apply advanced statistical techniques to analyze massive, diverse datasets to uncover insights that optimize our candidate evaluation processes and drive hiring excellence • Explore emerging technologies and innovative methodologies to enhance talent measurement while maintaining Amazon's commitment to scientific integrity • Translate complex research findings into compelling, actionable strategies that influence senior leader/business decisions and shape Amazon's talent acquisition roadmap • Write impactful documents that distill intricate scientific concepts into clear, persuasive communications for diverse audiences, from data scientists to business leaders • Ensure effective teamwork, communication, collaboration, and commitment across multiple teams with competing priorities A day in the life Imagine diving into challenges that impact millions of employees across Amazon's global operations. As a GHS Research Scientist, you'll tackle questions about hiring and organizational effectiveness on a global scale. Your day might begin with analyzing datasets to inform how we attract and select world-class talent. Throughout the day, you'll collaborate with peers in our research community, discussing different research methodologies and sharing innovative approaches to solving unique personnel challenges. This role offers a blend of focused analytical time and interacting with stakeholders across the globe.
Applied Scientist, AGI Information
US, WA, Seattle
We are looking for a researcher in state-of-the-art LLM technologies for applications across Alexa, AWS, and other Amazon businesses. In this role, you will innovate in the fastest-moving fields of current AI research, in particular in how to integrate a broad range of structured and unstructured information into AI systems (e.g. with RAG techniques), and get to immediately apply your results in highly visible Amazon products. If you are deeply familiar with LLMs, natural language processing, computer vision, and machine learning and thrive in a fast-paced environment, this may be the right opportunity for you. Our fast-paced environment requires a high degree of autonomy to deliver ambitious science innovations all the way to production. You will work with other science and engineering teams as well as business stakeholders to maximize velocity and impact of your deliverables. It's an exciting time to be a leader in AI research. In Amazon's AGI Information team, you can make your mark by improving information-driven experience of Amazon customers worldwide!
Economist, Prime Science
US, WA, Seattle
Amazon Prime is looking for an ambitious Economist to help create econometric insights for world-wide Prime. Prime is Amazon's premiere membership program, with over 200M members world-wide. This role is at the center of many major company decisions that impact Amazon's customers. These decisions span a variety of industries, each reflecting the diversity of Prime benefits. These range from fast-free e-commerce shipping, digital content (e.g., exclusive streaming video, music, gaming, photos), and grocery offerings. Prime Science creates insights that power these decisions. As an economist in this role, you will create statistical tools that embed causal interpretations. You will utilize massive data, state-of-the-art scientific computing, econometrics (causal, counterfactual/structural, time-series forecasting, experimentation), and machine-learning, to do so. Some of the science you create will be publishable in internal or external scientific journals and conferences. You will work closely with a team of economists, applied scientists, data professionals (business analysts, business intelligence engineers), product managers, and software engineers. You will create insights from descriptive statistics, as well as from novel statistical and econometric models. You will create internal-to-Amazon-facing automated scientific data products to power company decisions. You will write strategic documents explaining how senior company leaders should utilize these insights to create sustainable value for customers. These leaders will often include the senior-most leaders at Amazon. The team is unique in its exposure to company-wide strategies as well as senior leadership. It operates at the research frontier of utilizing data, econometrics, artificial intelligence, and machine-learning to form business strategies. A successful candidate will have demonstrated a capacity for building, estimating, and defending statistical models (e.g., causal, counterfactual, time-series, machine-learning) using software such as R, Python, or STATA. They will have a willingness to learn and apply a broad set of statistical and computational techniques to supplement deep-training in one area of econometrics. For example, many applications on the team use structural econometrics, machine-learning, and time-series forecasting. They rely on building scalable production software, which involves a broad set of world-class software-building skills often learned on-the-job. As a consequence, already-obtained knowledge of SQL, machine learning, and large-scale scientific computing using distributed computing infrastructures such as Spark-Scala or PySpark would be a plus. Additionally, this candidate will show a track-record of delivering projects well and on-time, preferably in collaboration with other team members (e.g. co-authors). Candidates must have very strong writing and emotional intelligence skills (for collaborative teamwork, often with colleagues in different functional roles), a growth mindset, and a capacity for dealing with a high-level of ambiguity. Endowed with these traits and on-the-job-growth, the role will provide the opportunity to have a large strategic, world-wide impact on the customer experiences of Prime members.
Senior Applied Scientist, Security Issue Management
US, WA, Seattle
We are open to hiring candidates to work out of one of the following locations: Seattle, WA, USA Are you interested in building Agentic AI solutions that solve complex builder experience challenges with significant global impact? The Security Tooling team designs and builds high-performance AI systems using LLMs and machine learning that identify builder bottlenecks, automate security workflows, and optimize the software development lifecycle—empowering engineering teams worldwide to ship secure code faster while maintaining the highest security standards. As a Senior Applied Scientist on our Security Tooling team, you will focus on building state-of-the-art ML models to enhance builder experience and productivity. You will identify builder bottlenecks and pain points across the software development lifecycle, design and apply experiments to study developer behavior, and measure the downstream impacts of security tooling on engineering velocity and code quality. Our team rewards curiosity while maintaining a laser-focus on bringing products to market that empower builders while maintaining security excellence. Competitive candidates are responsive, flexible, and able to succeed within an open, collaborative, entrepreneurial, startup-like environment. At the forefront of both academic and applied research in builder experience and security automation, you have the opportunity to work together with a diverse and talented team of scientists, engineers, and product managers and collaborate with other teams. This role offers a unique opportunity to work on projects that could fundamentally transform how builders interact with security tools and how organizations balance security requirements with developer productivity. Key job responsibilities • Design and implement novel AI/ML solutions for complex security challenges and improve builder experience • Drive advancements in machine learning and science • Balance theoretical knowledge with practical implementation • Navigate ambiguity and create clarity in early-stage product development • Collaborate with cross-functional teams while fostering innovation in a collaborative work environment to deliver impactful solutions • Design and execute experiments to evaluate the performance of different algorithms and models, and iterate quickly to improve results • Establish best practices for ML experimentation, evaluation, development and deployment You’ll need a strong background in AI/ML, proven leadership skills, and the ability to translate complex concepts into actionable plans. You’ll also need to effectively translate research findings into practical solutions. A day in the life • Integrate ML models into production security tooling with engineering teams • Build and refine ML models and LLM-based agentic systems that understand builder intent • Create agentic AI solutions that reduce security friction while maintaining high security standards • Prototype LLM-powered features that automate repetitive security tasks • Design and conduct experiments (A/B tests, observational studies) to measure downstream impacts of tooling changes on engineering productivity • Present experimental results and recommendations to leadership and cross-functional teams • Gather feedback from builder communities to validate hypotheses About the team Diverse Experiences Amazon Security values diverse experiences. Even if you do not meet all of the qualifications and skills listed in the job description, we encourage candidates to apply. If your career is just starting, hasn’t followed a traditional path, or includes alternative experiences, don’t let it stop you from applying. Why Amazon Security? At Amazon, security is central to maintaining customer trust and delivering delightful customer experiences. Our organization is responsible for creating and maintaining a high bar for security across all of Amazon’s products and services. We offer talented security professionals the chance to accelerate their careers with opportunities to build experience in a wide variety of areas including cloud, devices, retail, entertainment, healthcare, operations, and physical stores Inclusive Team Culture In Amazon Security, it’s in our nature to learn and be curious. Ongoing DEI events and learning experiences inspire us to continue learning and to embrace our uniqueness. Addressing the toughest security challenges requires that we seek out and celebrate a diversity of ideas, perspectives, and voices. Training & Career Growth We’re continuously raising our performance bar as we strive to become Earth’s Best Employer. That’s why you’ll find endless knowledge-sharing, training, and other career-advancing resources here to help you develop into a better-rounded professional. Work/Life Balance We value work-life harmony. Achieving success at work should never come at the expense of sacrifices at home, which is why flexible work hours and arrangements are part of our culture. When we feel supported in the workplace and at home, there’s nothing we can’t achieve.
Sr Manager, Applied Science, Last Mile Delivery Automation
US, WA, Bellevue
We are seeking a Senior Manager, Applied Science to lead the applied science charter for Amazon’s Last-Hundred-Yard automation initiative, developing the algorithms, models, and learning systems that enable safe, reliable, and scalable autonomous delivery from vehicle to customer doorstep. This role owns the scientific direction across perception, localization, prediction, planning, learning-based controls, human-robot interaction (HRI), and data-driven autonomy validation, operating in complex, unstructured real-world environments. The Senior Manager will build and lead a high-performing team of applied scientists, set the technical vision and research-to-production roadmap, and ensure tight integration between science, engineering, simulation, and operations. This leader is responsible for translating ambiguous real-world delivery problems into rigorous modeling approaches, measurable autonomy improvements, and production-ready solutions that scale across cities, terrains, weather conditions, and customer scenarios. Success in this role requires deep expertise in machine learning and robotics, strong people leadership, and the ability to balance long-term scientific innovation with near-term delivery milestones. The Senior Manager will play a critical role in defining how Amazon applies science to unlock autonomous last-mile delivery at scale, while maintaining the highest bars for safety, customer trust, and operational performance. Key job responsibilities Set and own the applied science vision and roadmap for last-hundred-yard automation, spanning perception, localization, prediction, planning, learning-based controls, and HRI. Build, lead, and develop a high-performing applied science organization, including hiring, mentoring, performance management, and technical bar-raising. Drive the end-to-end science lifecycle from problem formulation and data strategy to model development, evaluation, deployment, and iteration in production. Partner closely with autonomy engineering to translate scientific advances into scalable, production-ready autonomy behaviors. Define and own scientific success metrics (e.g., autonomy performance, safety indicators, scenario coverage, intervention reduction) and ensure measurable impact. Lead the development of learning-driven autonomy using real-world data, simulation, and offline/online evaluation frameworks. Establish principled approaches for generalization across environments, including weather, terrain, lighting, customer properties, and interaction scenarios. Drive alignment between real-world operations and simulation, ensuring tight feedback loops for data collection and model validation. Influence safety strategy and validation by defining scientific evidence required for autonomy readiness and scale. Represent applied science in executive reviews, articulating trade-offs, risks, and long-term innovation paths.
Applied Science Manager, Amazon Industrial Robotics
US, MA, N.reading
Amazon Industrial Robotics is seeking exceptional talent to help develop the next generation of advanced robotics systems that will transform automation at Amazon's scale. We're building revolutionary robotic systems that combine cutting-edge AI, sophisticated control systems, and advanced mechanical design to create adaptable automation solutions capable of working safely alongside humans in dynamic environments. This is a unique opportunity to shape the future of robotics and automation at unprecedented scale, working with world-class teams pushing the boundaries of what's possible in robotic manipulation, locomotion, and human-robot interaction. This role presents an opportunity to shape the future of robotics through innovative applications of deep learning and large language models. At Amazon Industrial Robotics we leverage advanced robotics, machine learning, and artificial intelligence to solve complex operational challenges at unprecedented scale. Our fleet of robots operates across hundreds of facilities worldwide, working in sophisticated coordination to fulfill our mission of customer excellence. We are pioneering the development of robotics foundation models that: - Enable unprecedented generalization across diverse tasks - Enable unprecedented robustness and reliability, industry-ready - Integrate multi-modal learning capabilities (visual, tactile, linguistic) - Accelerate skill acquisition through demonstration learning - Enhance robotic perception and environmental understanding - Streamline development processes through reusable capabilities The ideal candidate will contribute to research that bridges the gap between theoretical advancement and practical implementation in robotics. You will be part of a team that's revolutionizing how robots learn, adapt, and interact with their environment. Join us in building the next generation of intelligent robotics systems that will transform the future of automation and human-robot collaboration. Key job responsibilities As an Applied Science Manager in the Foundations Model team, you will: - Build and lead a team of scientists and developers responsible for foundation model development - Define the right ‘FM recipe’ to reach industry ready solutions - Define the right strategy to ensure fast and efficient development, combining state of the art methods, research and engineering. - Lead Model Development and Training: Designing and implementing the model architectures, training and fine tuning the foundation models using various datasets, and optimize the model performance through iterative experiments - Lead Data Management: Process and prepare training data, including data governance, provenance tracking, data quality checks and creating reusable data pipelines. - Lead Experimentation and Validation: Design and execute experiments to test model capabilities on the simulator and on the embodiment, validate performance across different scenarios, create a baseline and iteratively improve model performance. - Lead Code Development: Write clean, maintainable, well commented and documented code, contribute to training infrastructure, create tools for model evaluation and testing, and implement necessary APIs - Research: Stay current with latest developments in foundation models and robotics, assist in literature reviews and research documentation, prepare technical reports and presentations, and contribute to research discussions and brainstorming sessions. - Collaboration: Work closely with senior scientists, engineers, and leaders across multiple teams, participate in knowledge sharing, support integration efforts with robotics hardware teams, and help document best practices and methodologies.
Sr. Applied Scientist, FAR (Frontier AI & Robotics)
CA, QC, Montreal
Join the next revolution in robotics at Amazon's Frontier AI & Robotics team, where you'll work alongside world-renowned AI pioneers to push the boundaries of what's possible in robotic intelligence. As an Applied Scientist, you'll be at the forefront of developing breakthrough foundation models that enable robots to perceive, understand, and interact with the world in unprecedented ways. You'll drive independent research initiatives in areas such as perception, manipulation, scene understanding, sim2real transfer, multi-modal foundation models, and multi-task learning, designing novel algorithms that bridge the gap between state-of-the-art research and real-world deployment at Amazon scale. In this role, you'll balance innovative technical exploration with practical implementation, collaborating with platform teams to ensure your models and algorithms perform robustly in dynamic real-world environments. You'll have access to Amazon's vast computational resources, enabling you to tackle ambitious problems in areas like very large multi-modal robotic foundation models and efficient, promptable model architectures that can scale across diverse robotic applications. Key job responsibilities - Design and implement novel deep learning architectures that push the boundaries of what robots can understand and accomplish - Drive independent research initiatives in robotics foundation models, focusing on breakthrough approaches in perception, and manipulation, for example open-vocabulary panoptic scene understanding, scaling up multi-modal LLMs, sim2real/real2sim techniques, end-to-end vision-language-action models, efficient model inference, video tokenization - Lead technical projects from conceptualization through deployment, ensuring robust performance in production environments - Collaborate with platform teams to optimize and scale models for real-world applications - Contribute to the team's technical strategy and help shape our approach to next-generation robotics challenges A day in the life - Design and implement novel foundation model architectures, leveraging our extensive compute infrastructure to train and evaluate at scale - Collaborate with our world-class research team to solve complex technical challenges - Lead technical initiatives from conception to deployment, working closely with robotics engineers to integrate your solutions into production systems - Participate in technical discussions and brainstorming sessions with team leaders and fellow scientists - Leverage our massive compute cluster and extensive robotics infrastructure to rapidly prototype and validate new ideas - Transform theoretical insights into practical solutions that can handle the complexities of real-world robotics applications About the team At Frontier AI & Robotics, we're not just advancing robotics – we're reimagining it from the ground up. Our team is building the future of intelligent robotics through ground breaking foundation models and end-to-end learned systems. We tackle some of the most challenging problems in AI and robotics, from developing sophisticated perception systems to creating adaptive manipulation strategies that work in complex, real-world scenarios. What sets us apart is our unique combination of ambitious research vision and practical impact. We leverage Amazon's massive computational infrastructure and rich real-world datasets to train and deploy state-of-the-art foundation models. Our work spans the full spectrum of robotics intelligence – from multimodal perception using images, videos, and sensor data, to sophisticated manipulation strategies that can handle diverse real-world scenarios. We're building systems that don't just work in the lab, but scale to meet the demands of Amazon's global operations. Join us if you're excited about pushing the boundaries of what's possible in robotics, working with world-class researchers, and seeing your innovations deployed at unprecedented scale.
Applied Scientist II, Amazon Business 3P
IN, TS, Hyderabad
We're seeking an Applied Scientist to lead and innovate in applying advanced AI technologies that will reshape how businesses sell on Amazon. Our team is passionate about leveraging Machine Learning, GenAI, and Agentic AI to help B2B sellers optimize their operations and drive growth. Join Amazon Business 3P (Third Party - Sellers) - a rapidly growing global organization where we innovate at the intersection of AI technology and B2B commerce. We're reimagining how sellers reach and serve business customers, creating intelligent solutions that help them grow their B2B business on Amazon. From AI-powered Seller Central tools to smart business certifications, dynamic pricing capabilities, and advanced analytics, we're transforming how B2B selling happens. As an Applied Scientist II on our AB 3P Tech team, you'll drive the development and implementation of state-of-the-art algorithms and models for supervised fine-tuning and reinforcement learning. You'll work with highly technical, entrepreneurial teams to: - Design and implement AI models that power the B2B selling experience - Lead the development of GenAI products that can handle Amazon-scale use cases - Drive research and implementation of advanced algorithms for human feedback and complex reasoning - Make strategic AI technology decisions and mentor technical talent - Own critical AI systems spanning from Seller Central to Amazon Business detail pages Join us in shaping the future of B2B selling - we're building applied AI solutions that businesses love and trust for their day-to-day success. If you are scrappy and bias for action is your favorite Leadership Principle, you'll fit right in as we innovate across the seller experience to create significant impact in this fast-growing business. Key job responsibilities Key job responsibilities: - Collaborate with cross-functional teams of engineers, product managers, and scientists to identify and solve complex problems in Gen AI - Design and execute experiments to evaluate the performance of different algorithms and models, and iterate quickly to improve results - Think big about the arc of development of Gen AI over a multi-year horizon, and identify new opportunities to apply these technologies to solve real-world problems - Communicate results and insights to both technical and non-technical audiences About the team At Amazon Business Third Party (AB3P) Tech, we're revolutionizing B2B e-commerce by empowering sellers in the business marketplace. Our scope spans the complete B2B selling journey, from Seller Central to Amazon Business detail pages, cart, and checkout for merchant-fulfilled offers. Our entrepreneurial culture and global reach define us. We develop features across seller experience, delivery, certifications, fees, registration, and analytics, collaborating with worldwide teams and leveraging advanced AI technologies to continuously innovate. Working in true Day 1 spirit, we build next-generation solutions that shape the future of B2B commerce. Join us in building next-generation solutions that shape the future of B2B commerce.
Data Scientist, Prime Video
GB, London
Come build the future of entertainment with us. Are you interested in shaping the future of movies and television? Prime Video is a premium streaming service that offers customers a vast collection of TV shows and movies - all with the ease of finding what they love to watch in one place. We offer customers thousands of popular movies and TV shows including Amazon Originals and exclusive licensed content to exciting live sports events. Prime Video is a fast-paced, growth business - available in over 200 countries and territories worldwide. The Video Content Research team works in a dynamic environment where innovating on behalf of our customers is at the heart of everything we do. We are seeking a Data Scientist to develop scalable models that uncover key insights into how, why and when customers engage with Prime Video marketing. Key job responsibilities In this role you will work closely with business stakeholders and technical peers (data scientists, economists and engineers) to develop causal marketing measurement models, analyze experiments and investigate customer, marketing and content related factors that drive engagement with Prime Video. You will create mechanisms and infrastructure to deploy complex models and generate insights at scale. You will have the opportunity to work with large datasets, work with AWS to build and deploy machine learning models that impact Prime Video's marketing decisions. About the team The Video Content Research team uses machine learning, econometrics, and data science to optimize Amazon's marketing and content investments. We generate insights for Amazon's digital video strategy, partnering with finance, marketing, and content teams. We analyze customer behavior on Prime Video (marketing impressions, clicks on owned channels) to identify optimization opportunities.
See more jobs
View from space of a connected network around planet Earth representing the Internet of Things.
Get more from Amazon Science
Subscribe to our monthly newsletter

Amazon.com | Conditions of Use | Privacy | © 1996-2025 Amazon.com, Inc. or its affiliates

Social