Endel Deeper Focus
Endel, creator of personalized soundscapes that have attracted more than 2 million users, has partnered with pioneering electronic producer and DJ Plastikman (Richie Hawtin) to release their AI-powered soundscape "Deeper Focus". The new soundscape is now available to Alexa customers.
Credit: Endel

How Endel’s AI-powered Focus soundscapes earned the backing of neuroscience

A new study has found that when compared to curated playlists and silence, personalized AI soundscapes generated by Alexa Fund company Endel are more effective in helping people focus. 

(Editor’s Note: This is the second in a series of articles Amazon Science is publishing related to the science behind products and services from companies in which Amazon has invested. The Amazon Alexa Fund first invested in Endel in 2018, and in 2020 the Alexa Fund participated in Endel’s $5 million Series A financing round.)

Founded in 2018, Endel creates personalized soundscapes to help people focus, relax, and sleep. Built on the back of its patented technology, Endel Pacific, the artificial intelligence (AI)-powered service takes into account the individual conditions of each listener, such as their heartbeat or the amount of light present, to generate customized sounds that help improve their well being, the company says.  

Endel provides three primary soundscapes: Focus, Relax, and Sleep. Using Alexa for instance, each soundscape is able to extract key local data about individuals, such as time of day, weather, and the amount of natural light, to help generate sound environments that improve these key states.

With the aim of helping people sleep better, Endel started collaborating with SleepScore Labs in 2020, to improve the effectiveness and user experience of their sleep soundscapes. The company also made headlines when it collaborated with musician Grimes to create “AI Lullaby”, a custom-made sleep soundscape which was made available through the Endel skill, for Alexa. 

Now, with the recent publication of a new white paper, "Differences In The Effects On Human Focus Of Music Playlists And Personalized Soundscapes, As Measured By Brain Signals", Focus mode is in the scientific spotlight. The white paper examines what properties of sound affect human focus, validating the company’s existing approach, while providing a roadmap for future improvements to its custom soundscapes. 

Focus results

Published by Arctop, a data and AI technology company that has developed a pioneering brain decoding SaaS solution, the white paper used Endel’s personalized Focus soundscapes, alongside focus-themed playlists from popular streaming platforms, to see how they affected a cross-section of users as they performed everyday tasks.

The white paper, which was authored by Arctop's research and development team led by principal investigator Dan Furman, PhD, and first author Aia Haruvi, MSc, was supported by Warner Music, Sony, Endel, and Universal Music, who provided the company with sounds, data, and financial support to help advance the research. The report looked at users at home in their natural environment, recording and interpreting their brain signals to show how they reacted to the music. The aim: examine what, if any, impact the use of Endel soundscapes, popular curated playlists that have been optimized for focus purposes, or just pure silence, had on the ability of listeners to perform tasks.  

“From the very beginning it has been important to us to be rooted in science,” says Endel co-founder and CEO Oleg Stavitsky. Upon launching Endel with renowned composer and sound designer Dmitry Evgrafov, Stavitsky began looking for research papers to inform their work within functional music, especially as it relates to helping people focus. 

“The majority of white papers out there would only reference popular music like Queen, Bach… making straightforward comments like, sad music makes you feel sad,” said Stavitsky.” There was nothing out there for what we were trying to apply here, anything that would help us go deeper and ask, ‘How does your brain react to certain types of sounds?’” 

When Stavitsky and Evgrafov received preliminary results from Arctop, it helped validate how their soundscapes impact brain activity on a second-by-second basis. 

Endel Image 1
The goal of the Arctop study was to determine what properties of sound affect human focus the most. This diagram from the research paper demonstrates the framework for reverse correlation of time-series focus values with audio features. A is an example of a recorded brain signal, B is an audio segment from one of the songs, and C shows the audio features dynamics during 30 minutes of recordings.
Credit: Arctop

“This was gold for us,” Stavitsky says about the results. “It wasn’t just about validating what we had, it was about how it worked specifically. Arctop has this proprietary system that allows you to zoom into a song, and on a second-by-second basis say ‘here’s the progression and here’s the brain activity.’”

Arctop examined participants doing various tasks in their home or work environments. Listening to either Endel soundscapes, curated playlists, or just silence, each volunteer completed four, one-hour sessions, that included a set tasks, and followed by an activity of their own preference. As the participants performed their tasks, they were monitored using state-of-the-art technology to pick up on the brain’s impulses, tracking its responses to the audio in relation to their task. Through this analysis, the team devised a ‘focus coefficient’, based on input from a brain decoded data electroencephalograph headband, and additional survey data from the participants. 

The results demonstrated that participants listening to personalized soundscapes increased their focus significantly when compared to listening to music playlists, or silence.

Endel releases new Deeper Focus soundscape on Alexa

On April 30, Endel made its latest soundscape, Deeper Focus, available on Alexa. Endel partnered with with pioneering electronic producer and DJ Plastikman (Richie Hawtin) to release their collaborative AI-powered soundscape. Read more about the new soundscape here.

“When we set up this experiment we didn’t know what would happen,” Furman explained. “One of our main takeaways is that the personalization of soundscapes is really effective.” 

The approach to the research is also relatively new, Furman explains. 

“One thing we want to highlight is that the method we used is naturalistic neuroscience – outside of the lab, with no technicians present, no wires... It was a uniquely natural capture of data. Here people were able to work at home, and use their own tablet or phone, they wore regular headphones and a light headband only for the brain decoding, which was really novel they were able to experience the content exactly as they would in everyday life. Ultimately, we believe that context, gives more credence to our findings.

How focus works

Endel’s founders believe the study provides new information that the company can use to enhance its soundscapes. Unlike its other modes, Sleep and Relax, Focus is the only Endel soundscape to employ percussion. But it’s more complex than just adding a few beats.

“The tempo is closely tied to your heart rate, and can adjust based on your resting and active heart rate,” explains Evgrafov, who works closely with fellow Endel sound designer, Alexander Vasilenko, to bring its soundscapes to life. “The sounds are more active, have less reverb and are more nuanced. There is a very gentle balance that must be maintained with the rhythm, as the brain starts to block out rhythmic sounds after a time.” 

Endel Image 2
This diagram from the Arctop white paper shows the study's processing pipeline. Data acquisition included at home EEG recordings of four sessions, each with a different background audio stream. EEG processing included filtering the signal, feature extraction, and training machine learning models to map between brain features and reported focus. Obtaining the brain decoded focus dynamics enables comparison of focus levels during different types of audio streams.
Credit: Arctop

Thanks to the way that in which the data was collected, Endel can now assess how sound impacts customers’ responses on a second-by-second basis.

“We got a lot of information about structure, where we realized we had to build it up, and relax things. In those transitions, the curve [for the focus coefficient] goes up drastically,” Evgrafov explains. “It’s not about the amount of instruments, it’s the nature of change that provides the most impact. This isn’t something we could have worked out ourselves.”

Further areas of focus 

From its inception, Endel has taken a scientific approach to programming its technology, applying information and knowledge on psychoacoustics that was readily available online, while at the same time relying on Evgrafov’s musicianship and heuristic knowledge.

“We started getting more neuroscience data, and that was more important for us, but we couldn’t answer simple questions like, ‘What is focus?’, and ‘What makes something relaxing?’ Now, thanks to the report, we see how the entire structure of a song impacts brain functionality,” Evgrafov said. “There are other layers of knowledge as well, such as the acoustical and other sound treatments that are present in the very spectrum of the sound. These parameters can help us program our core technology.”

This is more thorough and goes deeper than anything that has been done before, specifically about how sounds affect your cognitive state when it comes to concentration.
Oleg Stavitsky, Endel CEO

Now, Endel is focused on taking its AI technology to the next level. 

“To me, what is important is how groundbreaking this is,” says Stavitsky. “This is more thorough and goes deeper than anything that has been done before, specifically about how sounds affect your cognitive state when it comes to concentration.”

Both Arctop and Endel see potential in further exploring additional factors that weren’t examined in the report, such as how personalized soundscapes can affect productivity, creativity, and wellbeing — states that can be directly associated with focusing. Using current Arctop technology for headphones, earbuds, AR/VR devices and the ‘focus coefficient,’ for example, Endel soundscapes can adapt in real-time to fit an individual user’s precise needs for focusing in the moment.   

“We believe personalized soundscapes are the new way to experience functional music,” Stavitsky says. “We see it as a new category of music — functional music — and within this field, Endel is a leader.”

 

Research areas

Related content

US, WA, Seattle
Economists in this role partner with business stakeholders to distill complex problems into testable economic questions and generate actionable insights. They collaborate with engineers and scientists to estimate models on large-scale data, design pilots, measure impact, and scale successful prototypes into improved policies and programs. They leverage AI tools to scale economic study for broader business impact. They communicate findings to business leaders, incorporate feedback, and deliver customer-centric solutions at scale.
US, NY, New York
Are you passionate about solving big problems from ground-up? Do you enjoy building new state-of-the-art products at internet scale? Come lead the innovation in this startup team, vertical ad products. This is a green field problem without a known answer or a pattern to follow. We have ambitious vision to simplify full funnel advertising solutions, at scale, with specialized agentic AI-powered models and diversify the demand to strategic verticals including finserv, autos, locals.. etc. We are seeking an experienced Applied Scientist to drive innovation in our Ads Foundational Model. In this individual contributor role, you will apply advanced machine learning techniques to improve advertiser performance and customer experience. Key job responsibilities As an Applied Scientist on this team, you will: 1. Develop and drive the science strategy for Ads Foundational Model (Ads-FM), aligning it with the program's objectives and overall business goals. 2. Identify high-impact opportunities within Ads-FM program and lead the ideation, planning, and execution of science initiatives to address them. 3. Build and deploy machine learning models using computer vision, natural language processing, and deep learning to evaluate and enhance ad effectiveness. 4. Develop algorithms that extract meaningful signals from image, video, and audio content to predict and improve customer engagement 5. Leverage Amazon's extensive data repository to create predictive models that generate actionable recommendations for more compelling ad creative 6. Collaborate with business leaders and cross-functional teams to implement ML-powered solutions 7. Contribute to the ML roadmap for the Ads-FM program through innovation and research.
US, WA, Seattle
This role will contribute to developing the Economics and Science products and services in the Fee domain, with specialization in supply chain systems and fees. Through the lens of economics, you will develop causal links for how Amazon, Sellers and Customers interact. You will be a key and senior scientist, advising Amazon leaders how to price our services. You will work on developing frameworks and scaleable, repeatable models supporting optimal pricing and policy in the two-sided marketplace that is central to Amazon's business. The pricing for Amazon services is complex. You will partner with science and technology teams across Amazon including Advertising, Supply Chain, Operations, Prime, Consumer Pricing, and Finance. We are looking for an experienced Principal Economist to improve our understanding of seller Economics, enhance our ability to estimate the causal impact of fees, and work with partner teams to design pricing policy changes. In this role, you will provide guidance to scientists to develop econometric models to influence our fee pricing worldwide. You will lead the development of causal models to help isolate the impact of fee and policy changes from other business actions, using experiments when possible, or observational data when not. Key job responsibilities The ideal candidate will have extensive Economics knowledge, demonstrated strength in practical and policy relevant structural econometrics, strong collaboration skills, proven ability to lead highly ambiguous and large projects, and a drive to deliver results. They will work closely with Economists, Data / Applied Scientists, Strategy Analysts, Data Engineers, and Product leads to integrate economic insights into policy and systems production. Familiarity with systems and services that constitute seller supply chains is a plus but not required. About the team The Stores Economics and Sciences team is a central science team that supports Amazon's Retail and Supply Chain leadership. We tackle some of Amazon's most challenging economics and machine learning problems, where our mandate is to impact the business on massive scale.
US, CA, San Diego
The Private Brands team is looking for a Research Scientist to join the team in building science solutions at scale. Our team applies Optimization, Machine Learning, Statistics, Causal Inference, and Econometrics/Economics to derive actionable insights about the complex economy of Amazon’s retail business and develop Statistical Models and Algorithms to drive strategic business decisions and improve operations. We are an interdisciplinary team of Scientists, Engineers, and Economists. Key job responsibilities You will work with business leaders, scientists, and economists to translate business and functional requirements into concrete deliverables, including the design, development, testing, and deployment of highly scalable optimization solutions and ML models. This is a unique, high visibility opportunity for someone who wants to have business impact, dive deep into large-scale problems, enable measurable actions on the consumer economy, and work closely with scientists and economists. As a Research Scientist, you bring business and industry context to science and technology decisions. You set the standard for scientific excellence and make decisions that affect the way we build and integrate algorithms. Your solutions are exemplary in terms of algorithm design, clarity, model structure, efficiency, and extensibility. You tackle intrinsically hard problems, acquiring expertise as needed. You decompose complex problems into straightforward solutions. We are particularly interested in candidates with experience in Operations Research and predictive models and working with distributed systems. Academic and/or practical background in Operations Research, Machine Learning and Reinforcement Learning are particularly relevant for this position. To know more about Amazon science, Please visit https://www.amazon.science
US, CA, Palo Alto
Alexa for Shopping (previously Rufus) is seeking a Senior Manager, Applied Science to lead multidisciplinary teams of Applied Scientists and Machine Learning Engineers building next-generation conversational AI and multi-agent systems powering customer-facing experiences at scale. This leader will drive both scientific innovation and execution across large language models (LLMs), agent orchestration, retrieval and grounding systems, evaluation frameworks, and scalable AI infrastructure. The role requires a combination of deep technical judgment, organizational leadership, product and engineering partnership, and operational excellence. The ideal candidate has a strong track record of building high-performing science and engineering teams, translating ambiguous business problems into scalable AI solutions, and delivering measurable customer impact through applied machine learning and generative AI technologies. Key job responsibilities - Lead and grow teams of Applied Scientists and Machine Learning Engineers working on conversational AI and multi-agent orchestration systems. - Define and drive technical strategy for large-scale generative AI systems, including LLM routing, prompting, grounding, memory, tool use, personalization, and response optimization. - Partner closely with Product, Engineering, and Tech leadership to align AI investments with long-term business and customer goals. - Drive end-to-end delivery of production AI systems balancing quality, latency, scalability, safety, and operational reliability. - Establish scientific and engineering best practices across experimentation, evaluation, model iteration, and production deployment. - Lead roadmap prioritization and execution across research innovation and product delivery timelines. - Build scalable evaluation methodologies and quality frameworks for multilingual and global customer experiences. - Mentor and develop technical leaders across both science and engineering disciplines. - Foster a high-performance culture centered on customer obsession, innovation, operational excellence, and strong cross-functional collaboration.
US, NY, New York
We are seeking a Human-Robot Interaction (HRI) Applied Scientist to develop cutting-edge interactions that make robots feel alive, personal, and fun. In this role, you will focus on verbal and non-verbal conversational systems, social dynamics, memory, and long-term relationship formation between robots, their environments, and the people they interact with. Your contributions will be essential in advancing robotics by enabling expressive, socially intelligent, and trustworthy interactions between robots and humans. Key job responsibilities - Develop interactive systems that leverage large language models, multimodal inputs and outputs, reinforcement learning from human feedback, or other advanced techniques to achieve fluid, engaging, and socially appropriate robot behavior - Design and implement intelligent conversational systems that handle turn-taking, grounding, interruption, and incorporates context drawn from a robot's physical environment and shared history with a user - Integrate perceptual sensor streams including gaze, facial expression, gesture, posture, and more to understand social context and produce coherent, lifelike interactions. - Develop memory and personalization systems that allow robots to form lasting relationships with individual users, learn their environments, and adapt their behavior over weeks and months - Stay updated on advancements in HRI, NLP, multimodal AI, and cognitive and social science to apply cutting-edge techniques to robot interaction challenges - Lead technical projects from conception through production deployment - Mentor junior scientists and engineers - Bridge research initiatives with practical engineering implementation
IN, KA, Bengaluru
Do you want to join an innovative team of scientists applying machine learning and advanced statistical techniques to protect Amazon customers and enable a trusted eCommerce experience? Are you excited about modeling terabytes of data and building state-of-the-art algorithms to solve complex, real-world fraud and risk challenges? Do you enjoy owning end-to-end machine learning problems, directly influencing customer experience and company profitability, while collaborating in a diverse, high-performing team? If so, the Amazon Buyer Risk Prevention (BRP) Machine Learning team may be the right fit for you. We are seeking an Applied Scientist to design, develop, and deploy advanced algorithmic systems that safeguard millions of transactions every day. In this role, you will independently drive model development from problem formulation to production deployment, build scalable ML solutions, and leverage emerging technologies—including Generative AI and LLMs—to enhance fraud detection and next-generation risk prevention systems. Key job responsibilities Own end-to-end development of machine learning models for large-scale risk management systems Analyze large volumes of historical and real-time data to identify fraud patterns and emerging risk trends Design, develop, validate, and deploy innovative models to production environments Apply GenAI/LLM technologies to automate risk evaluation and improve operational efficiency Collaborate closely with software engineering teams to implement scalable, real-time model solutions Partner with operations and business stakeholders to translate risk insights into measurable impact Establish scalable and automated processes for data analysis, model experimentation, validation, and monitoring Track model performance and business metrics; communicate insights clearly to technical and non-technical stakeholders Research and implement novel machine learning and statistical methodologies
IN, KA, Bengaluru
Do you want to join an innovative team applying machine learning and advanced statistical techniques to protect Amazon customers and enable a trusted eCommerce experience? Are you excited about working with large-scale datasets and developing models that solve real-world fraud and risk challenges? If so, the Amazon Buyer Risk Prevention (BRP) Machine Learning team may be the right fit for you. We are seeking an Applied Scientist to help develop scalable machine learning solutions that safeguard millions of transactions every day. In this role, you will partner with senior scientists and engineers to translate business problems into data-driven solutions, build and evaluate models, and contribute to next-generation risk prevention systems, including applications of Generative AI and LLM technologies. Key job responsibilities Apply machine learning and statistical techniques to build and improve risk management models Analyze large-scale historical data to identify risk patterns and emerging trends Develop, validate, and deploy innovative models under the guidance of senior scientists Experiment with emerging technologies, including GenAI/LLMs, to enhance automation and risk evaluation Collaborate closely with software engineers to implement models in real-time production systems Partner with operations and business teams to improve risk policies and operational efficiency Build scalable, automated pipelines for data analysis, model training, and validation Monitor model performance and provide clear reporting on key risk and business metrics Research and prototype new modeling approaches to improve system performance
IN, KA, Bengaluru
Do you want to join an innovative team of scientists applying machine learning and advanced statistical techniques to protect Amazon customers and enable a trusted eCommerce experience? Are you excited about modeling terabytes of data and building state-of-the-art algorithms to solve complex, real-world fraud and risk challenges? Do you enjoy owning end-to-end machine learning problems, directly influencing customer experience and company profitability, while collaborating in a diverse, high-performing team? If so, the Amazon Buyer Risk Prevention (BRP) Machine Learning team may be the right fit for you. We are seeking an Applied Scientist to design, develop, and deploy advanced algorithmic systems that safeguard millions of transactions every day. In this role, you will independently drive model development from problem formulation to production deployment, build scalable ML solutions, and leverage emerging technologies—including Generative AI and LLMs—to enhance fraud detection and next-generation risk prevention systems. Key job responsibilities Own end-to-end development of machine learning models for large-scale risk management systems Analyze large volumes of historical and real-time data to identify fraud patterns and emerging risk trends Design, develop, validate, and deploy innovative models to production environments Apply GenAI/LLM technologies to automate risk evaluation and improve operational efficiency Collaborate closely with software engineering teams to implement scalable, real-time model solutions Partner with operations and business stakeholders to translate risk insights into measurable impact Establish scalable and automated processes for data analysis, model experimentation, validation, and monitoring Track model performance and business metrics; communicate insights clearly to technical and non-technical stakeholders Research and implement novel machine learning and statistical methodologies
IN, KA, Bengaluru
Do you want to lead the development of advanced machine learning systems that protect millions of customers and power a trusted global eCommerce experience? Are you passionate about modeling terabytes of data, solving highly ambiguous fraud and risk challenges, and driving step-change improvements through scientific innovation? If so, the Amazon Buyer Risk Prevention (BRP) Machine Learning team may be the right place for you. We are seeking a Senior Applied Scientist to define and drive the scientific direction of large-scale risk management systems that safeguard millions of transactions every day. In this role, you will lead the design and deployment of advanced machine learning solutions, influence cross-team technical strategy, and leverage emerging technologies—including Generative AI and LLMs—to build next-generation risk prevention platforms. Key job responsibilities Lead the end-to-end scientific strategy for large-scale fraud and risk modeling initiatives Define problem statements, success metrics, and long-term modeling roadmaps in partnership with business and engineering leaders Design, develop, and deploy highly scalable machine learning systems in real-time production environments Drive innovation using advanced ML, deep learning, and GenAI/LLM technologies to automate and transform risk evaluation Influence system architecture and partner with engineering teams to ensure robust, scalable implementations Establish best practices for experimentation, model validation, monitoring, and lifecycle management Mentor and raise the technical bar for junior scientists through reviews, technical guidance, and thought leadership Communicate complex scientific insights clearly to senior leadership and cross-functional stakeholders Identify emerging scientific trends and translate them into impactful production solutions