Compression of Acoustic Event Detection Models With Quantized Distillation

2019
Download Copy BibTeX
Copy BibTeX
Acoustic Event Detection (AED), aiming at detecting categories of events based on audio signals, has found application in many intelligent systems. Recently deep neural network significantly advances this field and reduces detection errors to a large scale. However how to efficiently execute deep models in AED has received much less attention. Meanwhile state-of-the-art AED models are based on large deep models, which are computational demanding and challenging to deploy on devices with constrained computational resources. In this paper, we present a simple yet effective compression approach which jointly lever-ages knowledge distillation and quantization to compress larger network (teacher model) into compact network (student model). Experimental results show proposed technique not only lowers error rate of original compact network by15%through distillation but also further reduces its model size to a large extent (2%of teacher,12%of full-precision student) through quantization. Index Terms: acoustic event detection, model compression, quantization, knowledge distillation
Research areas
Tags

Latest news

US, CA, Palo Alto
We are looking for a passionate Applied Scientist to help pioneer the next generation of agentic AI applications for Amazon advertisers. In this role, you will design agentic architectures, develop tools and datasets, and contribute to building systems that can reason, plan, and act autonomously across complex advertiser workflows. You will work at the forefront of applied AI, developing methods for fine-tuning, reinforcement learning, and preference optimization, while helping create evaluation frameworks that ensure safety, reliability, and trust at scale. You will work backwards from the needs of advertisers—delivering customer-facing products that directly help them create, optimize, and grow their campaigns. Beyond building models, you will advance the agent ecosystem by experimenting with and applying core primitives such as tool orchestration, multi-step reasoning, and adaptive preference-driven behavior. This role requires working independently on ambiguous technical problems, collaborating closely with scientists, engineers, and product managers to bring innovative solutions into production. Key job responsibilities - Lead business, science and engineering strategy and roadmap for Sponsored Products Agentic Advertiser Guidance. - Design and build agents to guide advertisers in conversational and non-conversational experience. - Design and implement advanced model and agent optimization techniques, including supervised fine-tuning, instruction tuning and preference optimization (e.g., DPO/IPO). - Curate datasets and tools for MCP. - Build evaluation pipelines for agent workflows, including automated benchmarks, multi-step reasoning tests, and safety guardrails. - Develop agentic architectures (e.g., CoT, ToT, ReAct) that integrate planning, tool use, and long-horizon reasoning. - Prototype and iterate on multi-agent orchestration frameworks and workflows. - Collaborate with peers across engineering and product to bring scientific innovations into production. - Stay current with the latest research in LLMs, RL, and agent-based AI, and translate findings into practical applications. About the team The Sponsored Products and Brands team at Amazon Ads is re-imagining the advertising landscape through the latest generative AI technologies, revolutionizing how millions of customers discover products and engage with brands across Amazon.com and beyond. We are at the forefront of re-inventing advertising experiences, bridging human creativity with artificial intelligence to transform every aspect of the advertising lifecycle from ad creation and optimization to performance analysis and customer insights. We are a passionate group of innovators dedicated to developing responsible and intelligent AI technologies that balance the needs of advertisers, enhance the shopping experience, and strengthen the marketplace. If you're energized by solving complex challenges and pushing the boundaries of what's possible with AI, join us in shaping the future of advertising. The Advertiser Guidance team within Sponsored Products and Brands is focused on guiding and supporting 1.6MM advertisers to meet their advertising needs of creating and managing ad campaigns. At this scale, the complexity of diverse advertiser goals, campaign types, and market dynamics creates both a massive technical challenge and a transformative opportunity: even small improvements in guidance systems can have outsized impact on advertiser success and Amazon’s retail ecosystem. Our vision is to build a highly personalized, context-aware agentic advertiser guidance system that leverages LLMs together with tools such as auction simulations, ML models, and optimization algorithms. This agentic framework, will operate across both chat and non-chat experiences in the ad console, scaling to natural language queries as well as proactively delivering guidance based on deep understanding of the advertiser. To execute this vision, we collaborate closely with stakeholders across Ad Console, Sales, and Marketing to identify opportunities—from high-level product guidance down to granular keyword recommendations—and deliver them through a tailored, personalized experience. Our work is grounded in state-of-the-art agent architectures, tool integration, reasoning frameworks, and model customization approaches (including tuning, MCP, and preference optimization), ensuring our systems are both scalable and adaptive.
US, NY, New York
We are looking for a passionate Applied Scientist to help pioneer the next generation of agentic AI applications for Amazon advertisers. In this role, you will design agentic architectures, develop tools and datasets, and contribute to building systems that can reason, plan, and act autonomously across complex advertiser workflows. You will work at the forefront of applied AI, developing methods for fine-tuning, reinforcement learning, and preference optimization, while helping create evaluation frameworks that ensure safety, reliability, and trust at scale. You will work backwards from the needs of advertisers—delivering customer-facing products that directly help them create, optimize, and grow their campaigns. Beyond building models, you will advance the agent ecosystem by experimenting with and applying core primitives such as tool orchestration, multi-step reasoning, and adaptive preference-driven behavior. This role requires working independently on ambiguous technical problems, collaborating closely with scientists, engineers, and product managers to bring innovative solutions into production. Key job responsibilities - Design and build agents to guide advertisers in conversational and non-conversational experience. - Design and implement advanced model and agent optimization techniques, including supervised fine-tuning, instruction tuning and preference optimization (e.g., DPO/IPO). - Curate datasets and tools for MCP. - Build evaluation pipelines for agent workflows, including automated benchmarks, multi-step reasoning tests, and safety guardrails. - Develop agentic architectures (e.g., CoT, ToT, ReAct) that integrate planning, tool use, and long-horizon reasoning. - Prototype and iterate on multi-agent orchestration frameworks and workflows. - Collaborate with peers across engineering and product to bring scientific innovations into production. - Stay current with the latest research in LLMs, RL, and agent-based AI, and translate findings into practical applications. About the team The Sponsored Products and Brands team at Amazon Ads is re-imagining the advertising landscape through the latest generative AI technologies, revolutionizing how millions of customers discover products and engage with brands across Amazon.com and beyond. We are at the forefront of re-inventing advertising experiences, bridging human creativity with artificial intelligence to transform every aspect of the advertising lifecycle from ad creation and optimization to performance analysis and customer insights. We are a passionate group of innovators dedicated to developing responsible and intelligent AI technologies that balance the needs of advertisers, enhance the shopping experience, and strengthen the marketplace. If you're energized by solving complex challenges and pushing the boundaries of what's possible with AI, join us in shaping the future of advertising. The Advertiser Guidance team within Sponsored Products and Brands is focused on guiding and supporting 1.6MM advertisers to meet their advertising needs of creating and managing ad campaigns. At this scale, the complexity of diverse advertiser goals, campaign types, and market dynamics creates both a massive technical challenge and a transformative opportunity: even small improvements in guidance systems can have outsized impact on advertiser success and Amazon’s retail ecosystem. Our vision is to build a highly personalized, context-aware agentic advertiser guidance system that leverages LLMs together with tools such as auction simulations, ML models, and optimization algorithms. This agentic framework, will operate across both chat and non-chat experiences in the ad console, scaling to natural language queries as well as proactively delivering guidance based on deep understanding of the advertiser. To execute this vision, we collaborate closely with stakeholders across Ad Console, Sales, and Marketing to identify opportunities—from high-level product guidance down to granular keyword recommendations—and deliver them through a tailored, personalized experience. Our work is grounded in state-of-the-art agent architectures, tool integration, reasoning frameworks, and model customization approaches (including tuning, MCP, and preference optimization), ensuring our systems are both scalable and adaptive.
IL, Haifa
Come join the AWS Agentic AI science team in building the next generation models for intelligent automation. AWS, the world-leading provider of cloud services, has fostered the creation and growth of countless new businesses, and is a positive force for good. Our customers bring problems that will give Applied Scientists like you endless opportunities to see your research have a positive and immediate impact in the world. You will have the opportunity to partner with technology and business teams to solve real-world problems, have access to virtually endless data and computational resources, and to world-class engineers and developers that can help bring your ideas into the world. As part of the team, we expect that you will develop innovative solutions to hard problems, and publish your findings at peer reviewed conferences and workshops. Key job responsibilities We are looking for world class researchers with experience in one or more of the following areas - autonomous agents, API orchestration, Planning, large multimodal models (especially vision-language models), reinforcement learning (RL) and sequential decision making. A day in the life - Diverse Experiences Amazon values diverse experiences. Even if you do not meet all of the preferred 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 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. - 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 in the cloud. - 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 (gender diversity) conferences, inspire us to never stop embracing our uniqueness. - Mentorship and 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.
US, CA, Palo Alto
We are looking for a motivated Applied Scientist to join the team pioneering the next generation of agentic AI applications for Amazon advertisers. In this role, you will contribute to the design and development of agentic architectures, tools, and datasets that enable agents to reason, plan, and act autonomously across advertiser workflows. You will apply machine learning and large language model techniques—such as fine-tuning, reinforcement learning, and preference optimization—to solve real customer problems and improve advertiser outcomes at scale. You will work closely with senior scientists and engineers to experiment with new methods, run large-scale evaluations, and bring research ideas into production. You will be hands-on in implementing models, analyzing data, and building components that make our guidance agents more context-aware, reliable, and effective. Most importantly, you will work backwards from advertiser needs, contributing to customer-facing products that help advertisers create, optimize, and grow their campaigns. This is a highly collaborative and growth-oriented role, ideal for someone who thrives at the intersection of research and engineering, enjoys tackling ambiguous problems, and wants to shape the future of agent-based AI in advertising. Key job responsibilities - Contribute to the design and development of agents that guide advertisers across conversational and non-conversational experiences. - Implement and experiment with model and agent optimization techniques such as supervised fine-tuning and instruction tuning under the guidance of senior scientists. - Support dataset curation and tooling for model customization and preference optimization (e.g., MCP pipelines). - Build and maintain components of evaluation pipelines for agent workflows, including benchmark setup, automated test creation, and analysis of reasoning quality. - Prototype and validate elements of agentic architectures (e.g., CoT, ReAct, or ToT) to improve planning, reasoning, and tool use. - Conduct experiments, analyze performance, and communicate insights to drive iterative improvements to models and agents. - Collaborate with scientists, engineers, and product managers to integrate research outputs into production systems. - Stay current with emerging methods in LLMs, reinforcement learning, and agentic AI, and apply them to real-world advertiser scenarios. About the team The Sponsored Products and Brands team at Amazon Ads is re-imagining the advertising landscape through the latest generative AI technologies, revolutionizing how millions of customers discover products and engage with brands across Amazon.com and beyond. We are at the forefront of re-inventing advertising experiences, bridging human creativity with artificial intelligence to transform every aspect of the advertising lifecycle from ad creation and optimization to performance analysis and customer insights. We are a passionate group of innovators dedicated to developing responsible and intelligent AI technologies that balance the needs of advertisers, enhance the shopping experience, and strengthen the marketplace. If you're energized by solving complex challenges and pushing the boundaries of what's possible with AI, join us in shaping the future of advertising. The Advertiser Guidance team within Sponsored Products and Brands is focused on guiding and supporting 1.6MM advertisers to meet their advertising needs of creating and managing ad campaigns. At this scale, the complexity of diverse advertiser goals, campaign types, and market dynamics creates both a massive technical challenge and a transformative opportunity: even small improvements in guidance systems can have outsized impact on advertiser success and Amazon’s retail ecosystem. Our vision is to build a highly personalized, context-aware agentic advertiser guidance system that leverages LLMs together with tools such as auction simulations, ML models, and optimization algorithms. This agentic framework, will operate across both chat and non-chat experiences in the ad console, scaling to natural language queries as well as proactively delivering guidance based on deep understanding of the advertiser. To execute this vision, we collaborate closely with stakeholders across Ad Console, Sales, and Marketing to identify opportunities—from high-level product guidance down to granular keyword recommendations—and deliver them through a tailored, personalized experience. Our work is grounded in state-of-the-art agent architectures, tool integration, reasoning frameworks, and model customization approaches (including tuning, MCP, and preference optimization), ensuring our systems are both scalable and adaptive.
US, NY, New York
AMX Science's mission is to develop science that shapes human behavior in managing Amazon’s talent. We develop the core science for all Amazon-wide talent management and development experiences. Our multidisciplinary science team comprises of applied scientists, data scientists, economists and research scientists. We partner closely with product teams to build scalable science solutions that work backwards from internal customer problems for all of Amazon's businesses and locations around the world. Some of our work includes GenAI-powered writing assistance and insights, talent development and matching recommendations, experimentation and north star metrics, predictive and root cause models for talent events, voice of the customer qualitative analyses frameworks, and talent evaluation framework research. We are looking for an experienced AI/ML Applied Science Manager who has experience leading teams that build, apply and customize GenAI and traditional ML solutions to solve customer problems in production settings. Techniques we use on the team include NLP, supervised and unsupervised learning, recommendation systems, machine learning on graphs, reinforcement learning, algorithmic fairness and others on rich and novel datasets. As a science manager on the team, you will lead a team of ML scientists to build AI/ML solutions to address talent management and development product needs. You will be a hands-on technical leader who excels at driving innovation, fostering a data-driven culture, and leading through ambiguity to deliver measurable impact. You will innovate in the fastest-moving fields of current AI applications, including AI agents and intersection of GenAI and traditional ML systems, such as recommendations, and get to immediately apply your results in highly visible internal Amazon products that have a significant impact on employees’ lives. You will work closely with customers, product and program managers, other engineering managers, and tech leads to understand and guide your teams to build the right solutions. You will develop science roadmaps, communicate your vision and milestones to leadership and to your collaborators in the People Experience and Technology space. If this kind of work excites you, reach out to us to find out more! About the team AMX Science is an experienced central interdisciplinary organization of scientists spanning machine learning, economics and research that builds science models for Amazon's worldwide employee-facing talent management products, designs and supports experiments for product features, and measures impact of product and program initiatives across the broader organization. Examples of our work include GenAI-powered summarization and writing assistants, content and people recommendation systems, scalable experimentation products and measuring organizational north star metrics.
US, WA, Seattle
Calling all innovative tech enthusiasts! Join our cutting-edge team and dive into the world of distributed systems and high-performance computing. You'll have the opportunity to work on groundbreaking technologies that push the boundaries of computational science, solving complex challenges that have real-world impact. Are you passionate about creating scalable, sustainable computing systems that can power the world's most complex technological challenges? We're seeking innovative graduate researchers to push the boundaries of distributed systems and high-performance computing. We work across multiple Amazon businesses including Annapurna Labs, S3, EC2, and other critical infrastructure teams, though our research is not limited to these organizations. Our teams are committed to pushing the boundaries of distributed systems and high-performance computing, creating solutions that transform how we process and understand complex data Key job responsibilities • Collaborate with senior researchers to design and implement distributed computing solutions. • Design and prototype novel distributed computing architectures that enhance system performance and reliability • Conduct advanced research on scalable fault-tolerant systems for data center and serverless environments An ideal candidate for this role should possess a robust foundation in distributed systems, network architecture, or high-performance computing. The candidate should have hands-on experience with designing, implementing, and optimizing distributed algorithms, scalable network protocols, or parallel computing frameworks. Additionally, they must demonstrate the ability to work seamlessly within interdisciplinary teams, bringing together expertise from various domains such as software engineering, data science, and hardware architecture. This collaborative mindset is essential for developing innovative solutions that push the boundaries of cloud computing technology. A day in the life Your internship will be an immersive journey into advanced computational research. You'll collaborate with world-class scientists and engineers, exploring innovative approaches to solving complex computational problems. Expect to engage in hands-on projects that challenge your technical skills and spark your creativity.
US, CA, Sunnyvale
Prime Video is a first-stop entertainment destination offering customers a vast collection of premium programming in one app available across thousands of devices. Prime members can customize their viewing experience and find their favorite movies, series, documentaries, and live sports – including Amazon MGM Studios-produced series and movies; licensed fan favorites; and programming from Prime Video subscriptions such as Apple TV+, HBO Max, Peacock, Crunchyroll and MGM+. All customers, regardless of whether they have a Prime membership or not, can rent or buy titles via the Prime Video Store, and can enjoy even more content for free with ads. Are you interested in shaping the future of entertainment? Prime Video's technology teams are creating best-in-class digital video experience. As a Prime Video team member, you’ll have end-to-end ownership of the product, user experience, design, and technology required to deliver state-of-the-art experiences for our customers. You’ll get to work on projects that are fast-paced, challenging, and varied. You’ll also be able to experiment with new possibilities, take risks, and collaborate with remarkable people. We’ll look for you to bring your diverse perspectives, ideas, and skill-sets to make Prime Video even better for our customers. With global opportunities for talented technologists, you can decide where a career Prime Video Tech takes you! Key job responsibilities As an Applied Scientist at Prime Video, you will have end-to-end ownership of the product, related research and experimentation, applying advanced machine learning techniques in computer vision (CV), Generative AI, multimedia understanding and so on. You’ll work on diverse projects that enhance Prime Video’s content localization, image/video understanding, and content personalization, driving impactful innovations for our global audience. Other responsibilities include: - Research and develop generative models for controllable synthesis across images, video, vector graphics, and multimedia - Innovate in advanced diffusion and flow-based methods (e.g., inverse flow matching, parameter efficient training, guided sampling, test-time adaptation) to improve efficiency, controllability, and scalability. - Advance visual grounding, depth and 3D estimation, segmentation, and matting for integration into pre-visualization, compositing, VFX, and post-production pipelines. - Design multimodal GenAI workflows including visual-language model tooling, structured prompt orchestration, agentic pipelines. A day in the life Prime Video is pioneering the use of Generative AI to empower the next generation of creatives. Our mission is to make world-class media creation accessible, scalable, and efficient. We are seeking an Applied Scientist to advance the state of the art in Generative AI and to deliver these innovations as production-ready systems at Amazon scale. Your work will give creators unprecedented freedom and control while driving new efficiencies across Prime Video’s global content and marketing pipelines. This is a newly formed team within Prime Video Science!
US, NY, New York
The Artificial General Intelligence (AGI) team is looking for a passionate, talented, and inventive Applied Scientist to work on pre-training methodologies for Generative Artificial Intelligence (GenAI) models. You will interact closely with our customers and with the academic and research communities. Key job responsibilities Join us to work as an integral part of a team that has experience with GenAI models in this space. We work on these areas: - Scaling laws - Hardware-informed efficient model architecture, low-precision training - Optimization methods, learning objectives, curriculum design - Deep learning theories on efficient hyperparameter search and self-supervised learning - Learning objectives and reinforcement learning methods - Distributed training methods and solutions - AI-assisted research About the team The AGI team has a mission to push the envelope in GenAI with Large Language Models (LLMs) and multimodal systems, in order to provide the best-possible experience for our customers.
US, WA, Bellevue
Join Amazon's Customer Delivery Experience (CDE) Science Team as an Data Scientist to revolutionize global logistics through revolutionary research and innovation. Our team combines advanced machine learning and transportation logistics to improve delivery experiences for millions of customers worldwide. Working collaboratively with Amazon's core logistics teams, you will translate scientific innovations into practical solutions that drive meaningful improvements across our global fulfillment and delivery network. Key job responsibilities - Develop predictive algorithms to prevent delivery disruptions - Design machine learning systems for network-wide decision-making - Create mathematical models that optimize delivery speed, cost, and reliability - Implement generative AI solutions to enhance customer experiences - Produce research reports and contribute to peer-reviewed publications - Collaborate across teams to develop scalable, innovative solutions A day in the life Imagine a role where your analytical skills directly impact customer satisfaction. You'll dive into complex datasets, collaborate with cross-functional teams, and translate sophisticated research into practical logistics solutions. Your work bridges academic rigor with tangible business improvements, seeing your innovations come to life in real-time. About the team We are a dynamic, multidisciplinary group passionate about pushing the boundaries of supply chain science. Our team combines advanced analytics with operational excellence, creating innovative solutions that transform how millions of customers receive their packages. We value diverse perspectives and collaborative problem-solving, offering an environment where scientific creativity meets practical impact.
US, WA, Seattle
Are you a passionate Applied Scientist (AS) ready to shape the future of digital content creation? At Amazon, we're building Earth's most desired destination for creators to monetize their unique skills, inspire the next generation of customers, and help brands expand their reach. We build innovative products and experiences that drive growth for creators across Amazon's ecosystem. Our team owns the entire Creator product suite, ensuring a cohesive experience, optimizing compensation structures, and launching features that help creators achieve both monetary and non-monetary goals. Key job responsibilities As an AS on our team, you will: Handle challenging problems that directly impact millions of creators and customers Independently collect and analyze data Develop and deliver scalable predictive models, using any necessary programming, machine learning, and statistical analysis software Collaborate with other scientists, engineers, product managers, and business teams to creatively solve problems, measure and estimate risks, and constructively critique peer research Consult with engineering teams to design data and modeling pipelines which successfully interface with new and existing software Participate in design and implementation across teams to contribute to initiatives and develop optimal solutions that benefit the creators organization Key job responsibilities he successful candidate is a self-starter, comfortable with a dynamic, fast-paced environment, and able to think big while paying careful attention to detail. You have deep knowledge of an area/multiple areas of science, with a track record of applying this knowledge to deliver science solutions in a business setting and a demonstrated ability to operate at scale. You excel in a culture of invention and collaboration.