On-device speech processing makes Alexa faster, lower-bandwidth

Innovative training methods and model compression techniques combine with clever engineering to keep speech processing local.

At Amazon, we always look to invent new technology for improving customer experience. One technology we have been working on at Alexa is on-device speech processing, which has multiple benefits: a reduction in latency, or the time it takes Alexa to respond to queries; lowered bandwidth consumption, which is important on portable devices; and increased availability in in-car units and other applications where Internet connectivity is intermittent. On-device processing also enables the fusion of the speech signal with other modalities, like vision, for features such as Alexa’s natural turn-taking.

In the last year, we’ve continued to build upon Alexa’s on-device speech-processing capabilities. As a result of these inventions, we are launching a new setting that gives customers the option of having the audio of their Alexa voice requests processed locally, without being sent to the cloud.

In the cloud, storage space and computational capacity are effectively unconstrained. To ensure accuracy, our cloud models can be large and computationally demanding. Executing the same functions on-device means compressing our models into less than 1% as much space — with minimal loss in accuracy.

Moreover, in the cloud, the separate components of Alexa’s speech-processing stack — automatic speech recognition (ASR), whisper detection, and speaker identification — run on separate server nodes with their own powerful processors. On-device, those functions have to share hardware not only with each other but with Alexa’s other core device functions, such as music playback.

Re-creating Alexa’s speech-processing stack on-device was a massive undertaking. New methods for training small-footprint ASR models were part of the solution, but so were innovations in system design and hardware-software codesign. It was a joint effort across science and engineering teams over a span of years. Here’s a quick overview of how it works.

System architecture

Our on-device ASR model takes in an acoustic speech signal and outputs a set of hypotheses about what the speaker said, ranked according to probability. We represent those hypotheses as a lattice — a graph whose edges represent recognized words and the probability that a given word follows from the previous one.

Sample lattice.cropped.png
An example of a lattice representing ASR hypotheses.

With cloud-based ASR, encrypted audio streams to the cloud in small snippets called “frames”. With on-device ASR, only the lattice is sent to the cloud, where a large and powerful neural language model reranks the hypotheses. The lattice can’t be sent until the customer has finished speaking, as words later in a sequence can dramatically change the overall probability of a hypothesis.

The model that determines when the customer has finished speaking is called an end-pointer. End-pointers offer a natural trade-off between accuracy and latency: an aggressive end-pointer will initiate speech processing earlier, but it might cut the speaker off prematurely, resulting in a poor customer experience.

On the device, we in fact run two end-pointers: One is a speculative end-pointer that we have tuned to be about 200 milliseconds faster than the final end-pointer, so we can initiate downstream processing — such as natural-language understanding (NLU) — ahead of the final end-pointed ASR result. In exchange for speed, however, we trade off a little accuracy.

The final end-pointer takes longer to make a decision but is more accurate. In cases in which the first end-pointer cuts speech off too early, the final end-pointer sends a revised lattice and the instruction to reset downstream processing. In the large majority of cases, however, the aggressive end-pointer is correct, which reduces user-perceived latency, since downstream tasks are initiated earlier.

Another aspect of ASR that had to move on-device is context awareness. When computing the probabilities in a lattice, the ASR model should, for instance, give added weight to otherwise uncommon names that happen to be in the customer’s address book or the names the customer has assigned to household devices.

AmazonScience_StaticGraphic
A diagram of the on-device ASR network, with a closeup of the biasing mechanism that allows the network to ingest dynamic content. (Based on figures in "Context-aware Transformer transducer for speech recognition")
Attention map.png
This attention map indicates that the trained network is attending to the correct entry in a list of Alexa-linked home appliances. (From "Context-aware Transformer transducer for speech recognition")

Context awareness can’t wait for the cloud because the lattice, though it encodes multiple hypotheses, doesn’t come close to encoding all possible hypotheses. When constructing the lattice, the ASR system has to prune a lot of low-probability hypotheses. If context awareness isn’t built into the on-device model, names of contacts or linked skills might end up getting pruned.

Initially, we use a so-called shallow-fusion model to add context and personalize content on-device. When the system is building the lattice, it boosts the probabilities of contextually relevant words such as contact or appliance names.

The probability boosts are heuristic, however — they’re not learned jointly with the core ASR model. To achieve even better accuracy on personalized and long-tail content, we have developed a multihead attention-based context-biasing mechanism that is jointly trained with the rest of the ASR subnetworks.

Model training

On-device ASR required us to build a new model from the ground up, an end-to-end recurrent neural network-transducer (RNN-T) model that directly maps the input speech signal to an output sequence of words. Using a single neural network results in a significantly reduced memory footprint. But we had to develop new techniques, both for inference and for training, to achieve the degree of accuracy and compression that would let this technology handle utterances on-device.

Previously on Amazon Science, we’ve discussed some of the techniques we used to increase the accuracy of small-footprint end-to-end ASR models. With teacher-student training, for instance, we teach a small, lean model to match the outputs of a more-powerful but slower model. We developed a training methodology that made it possible to do teacher-student training efficiently with a million hours of unannotated speech.

Stream-level context.png
During the training of a context-aware ASR model, a long-short-term-memory (LSTM) encoder encodes both unlabeled and labeled segments of the audio stream, so the model can use the entire input audio to improve ASR accuracy. (From "Improving RNN-T ASR accuracy using context audio")

To further boost the accuracy of on-device RNN-T ASR, we developed techniques that allow the neural network to learn and exploit audio context within a stream. For example, for a stream comprising two utterances, “Alexa” and “Play a song”, the audio context from the keyword segment (“Alexa”) helps the model focus on the foreground speech and speaker. Separately, we implemented a novel discriminative-loss and training algorithm that aims at directly minimizing the word error rate (WER) of RNN-T ASR.

On top of these innovations, however, we still had to develop some new compression techniques to get the RNN-T to run efficiently on-device. A neural network consists of simple processing nodes each of which is connected to several others. The connections between nodes have associated weights, which determine how much one node’s output contributes to the computation performed by the next node.

One way to shrink a neural network’s memory footprint is to quantize its weights — to divide the total range of weights into a small set of intervals and use a single value to represent all the weights in each interval. So, for instance, the weights 0.70, 0.76, and 0.79 might all get quantized to the single value 0.75. Specifying an interval requires fewer bits than specifying several different floating-point values.

If quantization is done after a network has been trained, performance can suffer. We developed a method of <i class="rte2-style-italic">quantization-aware</i> training that imposes a probability distribution on the network weights during training, so that they can be easily quantized with little effect on performance. Unlike previous quantization-aware training methods, which mostly take quantization into account in the forward pass, ours accounts for quantization in the backward direction, during weight updates, through network loss regularization. And it does that efficiently.

A way to make neural networks run more efficiently — also a vital concern on resource-constrained devices — is to reduce low weights to zero. Computations involving zero weights can be discarded, reducing the computational burden.

Sparsification.png
Over successive training epochs, sparsification gradually drops low weights in a weight matrix.

But again, doing that reduction after the network is trained can compromise performance. We developed a <i class="rte2-style-italic">sparsification</i> method that enables the gradual reduction of low-value weights during training, so the network learns a model amenable to weight pruning.

Neural networks are typically trained on multiple passes through the same set of training data, or epochs. During each epoch, we force the network weights to diverge more and more, so that at the end of the final epoch, a fixed number of weights — say, half — are effectively zero. They can be safely discarded.

AmazonScience_AmnetDemo_V1.gif
A demonstration of the branching encoder network.

To improve on-device efficiency, we also developed a branching encoder network that uses two different neural networks to convert speech inputs into numeric representations suitable for speech classification. One network is complex, one simple, and the ASR model decides on the fly whether it can get away with passing an input frame to the simple model, saving computational cost and time. We described this work in more detail in an earlier Amazon Science blog post.

Hardware-software codesign

Quantization and sparsification make no difference to performance if the underlying hardware can’t take advantage of them. Another key to getting ASR to run on-device was the design of Amazon’s AZ family of neural edge processors, which are optimized for our specific approach to compression.

For one thing, where a typical processor might represent data using 16 or 32 bits, for certain core operations, the AZ processors accelerate computation by using an 8-bit or even lower-bit representation, because that’s all we need to handle quantized values.

The weights of a neural network are typically represented using a matrix — a big grid of numbers. A matrix half of whose values are zeroes takes up as much space as a matrix that’s all nonzero.

On computer chips, transferring data tends to be much more time consuming than executing computations. So when we load our matrix into memory, we use a compression scheme that takes advantage of low-bit quantization and zero values. The circuitry for decoding the compressed representation is built into the chip.

In the neural processor’s memory, the matrix is reconstituted: the zeroes are filled back in. But the processor’s circuitry is designed to recognize zero values and discard computations involving them. So the time savings from sparsification are realized in the hardware itself.

Moving speech recognition on device entails a number of innovations in other areas, such as reduction in the bandwidth required for model updates and compression of NLU models, to ensure basic functionality on devices with intermittent Internet connectivity. And we’re also hard at work on multilingual on-device ASR models for dynamic language switching, or automatically recognizing which of two languages a customer is speaking and responding in kind.

The launch of on-device speech processing is a huge step in bringing the benefits of “processing on the edge” to our customers, and we will continue to invent on their behalf in this area.

Related content

US, WA, Seattle
We are a team of doers working passionately to apply cutting-edge advances in deep learning in the life sciences to solve real-world problems. As a Senior Applied Science Manager you will participate in developing exciting products for customers. Our team rewards curiosity while maintaining a laser-focus in bringing products to market. Competitive candidates are responsive, flexible, and able to succeed within an open, collaborative, entrepreneurial, startup-like environment. At the leading edge of both academic and applied research in this product area, you have the opportunity to work together with a diverse and talented team of scientists, engineers, and product managers and collaborate with others teams. Location is in Seattle, US Embrace Diversity Here at Amazon, we embrace our differences. We are committed to furthering our culture of inclusion. We have ten employee-led affinity groups, reaching 40,000 employees in over 190 chapters globally. We have innovative benefit offerings, and host annual and ongoing learning experiences, including our Conversations on Race and Ethnicity (CORE) and AmazeCon (gender diversity) conferences. Amazon’s culture of inclusion is reinforced within our 14 Leadership Principles, which remind team members to seek diverse perspectives, learn and be curious, and earn trust Balance Work and Life Our team puts a high value on work-life balance. It isn’t about how many hours you spend at home or at work; it’s about the flow you establish that brings energy to both parts of your life. We believe striking the right balance between your personal and professional life is critical to life-long happiness and fulfillment. We offer flexibility in working hours and encourage you to find your own balance between your work and personal lives Mentor & Grow Careers Our team is dedicated to supporting new members. We have a broad mix of experience levels and tenures, and we’re building an environment that celebrates knowledge sharing and mentorship. Our senior members enjoy one-on-one mentoring and thorough, but kind, code reviews. We care about your career growth and strive to assign projects based on what will help each team member develop into a better-rounded engineer and enable them to take on more complex tasks in the future. Key job responsibilities • Manage high performing engineering and science teams • Hire and develop top-performing engineers, scientists, and other managers • Develop and execute on project plans and delivery commitments • Work with business, data science, software engineer, biological, and product leaders to help define product requirements and with managers, scientists, and engineers to execute on them • Build and maintain world-class customer experience and operational excellence for your deliverables
US, Virtual
The Amazon Economics Team is hiring Interns in Economics. We are looking for detail-oriented, organized, and responsible individuals who are eager to learn how to work with large and complicated data sets. Some knowledge of econometrics, as well as basic familiarity with Stata, R, or Python is necessary. Experience with SQL, UNIX, Sawtooth, and Spark would be a plus. These are full-time positions at 40 hours per week, with compensation being awarded on an hourly basis. You will learn how to build data sets and perform applied econometric analysis at Internet speed collaborating with economists, data scientists and MBAʼs. These skills will translate well into writing applied chapters in your dissertation and provide you with work experience that may help you with placement. Roughly 85% of interns from previous cohorts have converted to full time economics employment at Amazon. If you are interested, please send your CV to our mailing list at econ-internship@amazon.com.
US, WA, Seattle
Amazon internships are full-time (40 hours/week) for 12 consecutive weeks with start dates in May - July 2023. Our internship program provides hands-on learning and building experiences for students who are interested in a career in hardware engineering. This role will be based in Seattle, and candidates must be willing to work in-person. Corporate Projects (CPT) is a team that sits within the broader Corporate Development organization at Amazon. We seek to bring net-new, strategic projects to life by working together with customers and evolving projects from ZERO-to-ONE. To do so, we deploy our resources towards proofs-of-concept (POCs) and pilot programs and develop them from high-level ideas (the ZERO) to tangible short-term results that provide validating signal and a path to scale (the ONE). We work with our customers to develop and create net-new opportunities by relentlessly scouring all of Amazon and finding new and innovative ways to strengthen and/or accelerate the Amazon Flywheel. CPT seeks an Applied Science intern to work with a diverse, cross-functional team to build new, innovative customer experiences. Within CPT, you will apply both traditional and novel scientific approaches to solve and scale problems and solutions. We are a team where science meets application. A successful candidate will be a self-starter comfortable with ambiguity, strong attention to detail, and the ability to work in a fast-paced, ever-changing environment. As an Applied Science Intern, you will own the design and development of end-to-end systems. You’ll have the opportunity to create technical roadmaps, and drive production level projects that will support Amazon Science. You will work closely with Amazon scientists, and other science interns to develop solutions and deploy them into production. The ideal scientist must have the ability to work with diverse groups of people and cross-functional teams to solve complex business problems.
US, WA, Seattle
Amazon internships are full-time (40 hours/week) for 12 consecutive weeks with start dates in May - July 2023. Our internship program provides hands-on learning and building experiences for students who are interested in a career in hardware engineering. This role will be based in Seattle, and candidates must be willing to work in-person. Corporate Projects (CPT) is a team that sits within the broader Corporate Development organization at Amazon. We seek to bring net-new, strategic projects to life by working together with customers and evolving projects from ZERO-to-ONE. To do so, we deploy our resources towards proofs-of-concept (POCs) and pilot programs and develop them from high-level ideas (the ZERO) to tangible short-term results that provide validating signal and a path to scale (the ONE). We work with our customers to develop and create net-new opportunities by relentlessly scouring all of Amazon and finding new and innovative ways to strengthen and/or accelerate the Amazon Flywheel. CPT seeks an Applied Science intern to work with a diverse, cross-functional team to build new, innovative customer experiences. Within CPT, you will apply both traditional and novel scientific approaches to solve and scale problems and solutions. We are a team where science meets application. A successful candidate will be a self-starter comfortable with ambiguity, strong attention to detail, and the ability to work in a fast-paced, ever-changing environment. As an Applied Science Intern, you will own the design and development of end-to-end systems. You’ll have the opportunity to create technical roadmaps, and drive production level projects that will support Amazon Science. You will work closely with Amazon scientists, and other science interns to develop solutions and deploy them into production. The ideal scientist must have the ability to work with diverse groups of people and cross-functional teams to solve complex business problems.
US, CA, Palo Alto
The Amazon Search team creates powerful, customer-focused search solutions and technologies. Whenever a customer visits an Amazon site worldwide and types in a query or browses through product categories, Amazon Search services go to work. We design, develop, and deploy high performance, fault-tolerant distributed search systems used by millions of Amazon customers every day. We’re seeking a Principal Scientist with a deep expertise in Search Science. Your responsibilities will include everything from developing and prototyping innovative machine learning, and deep learning algorithms to implementing, testing, and supporting full solutions in a production environment. We are looking for innovators who can contribute to advancing search technology on what’s scientifically possible while remaining committed to creating world-class products. Joining this team, you’ll experience the benefits of working in a dynamic, entrepreneurial environment, while leveraging the resources of Amazon.com (AMZN), Earth's most customer-centric company one of the world's leading internet companies. We provide a highly customer-centric, team-oriented environment in our offices located in Palo Alto, California. Key job responsibilities As a hands-on leader of this team, you’ll be responsible for defining key research questions, identifying relevant data, adopting or proposing innovative machine learning solutions conducting rigorous experiments, publishing results and working with the engineering team to deploy these solutions. As a strategic leader, you will identify investment opportunities, develop long term strategies, and propose, prioritize and deliver on goals. You’ll also participate in organizational planning, hiring, mentorship and leadership development. You will be technically fearless and with a passion for building scalable science and engineering solutions. You will serve as a key scientific resource in full-cycle development (conception, design, implementation, testing to documentation, delivery, and maintenance). About the team Starting in 2009, the Visual Search & Augmented Reality team has thus far launched many visual search solutions on the Amazon App that use computer vision and machine learning/deep learning to help customers complete their shopping missions more easily; multiple internal teams at Amazon (devices, Kindle, Seller services, etc.) also use our libraries and APIs to deliver solutions to their own customers. We are a full stack shop, and our team capabilities cover the whole solution spectrum, ranging across applied science, large scale engineering services, product management, UX design, and mobile app development for iOS and Android.
US, MN, Minneapolis
AWS Central Economics is an interdisciplinary team on the cutting edge of economics, statistical analysis, and machine learning whose mission is to solve problems that have high risk with abnormally high returns. Our team leverages the strengths of our scientists to build solutions for some of the toughest business problems here at Amazon AWS. We are looking for an exceptionally talented, seasoned, and motivated Economist to manage a team of economists and data scientists to drive the science for AWS. Key job responsibilities Manage a team of economists and data scientists to deliver actionable economic analyses to business leaders, provide leadership on the economics and science used in the analyses, and engage with business leaders to identify challenges AWS faces that call for in-depth economic analyses and to ensure the analyses have their intended impact.
LU, Luxembourg
&ltHire Relocation Requisition - not for posting> Provides insights to leadership on improving Supply Chain cost and Speed by using Data Science and Analytics techniques. Build Dashboards and models to industrialize these findings at scale.
US, VA, Arlington
The People eXperience and Technology Central Science Team (PXTCS) uses economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements which simultaneously improve Amazon and the lives, wellbeing, and the value of work to Amazonians. We are an interdisciplinary team that combines the talents of science and engineering to develop and deliver solutions that measurably achieve this goal. We are looking for economists who are able to work with business partners to hone complex problems into specific, scientific questions, and test those questions to generate insights. The ideal candidate will work with engineers and computer scientists to estimate models and algorithms on large scale data, design pilots and measure their impact, and transform successful prototypes into improved policies and programs at scale. We are looking for creative thinkers who can combine a strong technical economic toolbox with a desire to learn from other disciplines, and who know how to execute and deliver on big ideas as part of an interdisciplinary technical team. Ideal candidates will work closely with business partners to develop science that solves the most important business challenges. They will work in a team setting with individuals from diverse disciplines and backgrounds. They will serve as an ambassador for science and a scientific resource for business teams, so that scientific processes permeate throughout the HR organization to the benefit of Amazonians and Amazon. Ideal candidates will own the data analysis, modeling, and experimentation that is necessary for estimating and validating models. They will work closely with engineering teams to develop scalable data resources to support rapid insights, and take successful models and findings into production as new products and services. They will be customer-centric and will communicate scientific approaches and findings to business leaders, listening to and incorporate their feedback, and delivering successful scientific solutions. Key job responsibilities Use causal inference methods to evaluate the impact of policies on employee outcomes. Examine how external labor market and economic conditions impact Amazon's ability to hire and retain talent. Use scientifically rigorous methods to develop and recommend career paths for employees. A day in the life Work with teammates to apply economic methods to business problems. This might include identifying the appropriate research questions, writing code to implement a DID analysis or estimate a structural model, or writing and presenting a document with findings to business leaders. Our economists also collaborate with partner teams throughout the process, from understanding their challenges, to developing a research agenda that will address those challenges, to help them implement solutions. About the team We are a multidisciplinary team that combines the talents of science and engineering to develop innovative solutions to make Amazon Earth's Best Employer.
US, WA, Seattle
The People eXperience and Technology Central Science Team (PXTCS) uses economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements which simultaneously improve Amazon and the lives, wellbeing, and the value of work to Amazonians. We are an interdisciplinary team that combines the talents of science and engineering to develop and deliver solutions that measurably achieve this goal. We are looking for economists who are able to apply economic methods to address business problems. The ideal candidate will work with engineers and computer scientists to estimate models and algorithms on large scale data, design pilots and measure their impact, and transform successful prototypes into improved policies and programs at scale. We are looking for creative thinkers who can combine a strong technical economic toolbox with a desire to learn from other disciplines, and who know how to execute and deliver on big ideas as part of an interdisciplinary technical team. Ideal candidates will work in a team setting with individuals from diverse disciplines and backgrounds. They will work with teammates to develop scientific models and conduct the data analysis, modeling, and experimentation that is necessary for estimating and validating models. They will work closely with engineering teams to develop scalable data resources to support rapid insights, and take successful models and findings into production as new products and services. They will be customer-centric and will communicate scientific approaches and findings to business leaders, listening to and incorporate their feedback, and delivering successful scientific solutions. Key job responsibilities Use causal inference methods to evaluate the impact of policies on employee outcomes. Examine how external labor market and economic conditions impact Amazon's ability to hire and retain talent. Use scientifically rigorous methods to develop and recommend career paths for employees. A day in the life Work with teammates to apply economic methods to business problems. This might include identifying the appropriate research questions, writing code to implement a DID analysis or estimate a structural model, or writing and presenting a document with findings to business leaders. Our economists also collaborate with partner teams throughout the process, from understanding their challenges, to developing a research agenda that will address those challenges, to help them implement solutions. About the team We are a multidisciplinary team that combines the talents of science and engineering to develop innovative solutions to make Amazon Earth's Best Employer.
US, WA, Seattle
Amazon is looking for talented Postdoctoral Scientists to join our global Science teams for a one-year, full-time research position. Postdoctoral Scientists will innovate as members of Amazon’s key global Science teams, including: AWS, Alexa AI, Alexa Shopping, Amazon Style, CoreAI, Last Mile, and Supply Chain Optimization Technologies. Postdoctoral Scientists will join one of may central, global science teams focused on solving research-intense business problems by leveraging Machine Learning, Econometrics, Statistics, and Data Science. Postdoctoral Scientists will work at the intersection of ML and systems to solve practical data driven optimization problems at Amazon scale. Postdocs will raise the scientific bar across Amazon by diving deep into exploratory areas of research to enhance the customer experience and improve efficiencies. Please note: This posting is one of several Amazon Postdoctoral Scientist postings. Please only apply to a maximum of 2 Amazon Postdoctoral Scientist postings that are relevant to your technical field and subject matter expertise. Key job responsibilities * Work closely with a senior science advisor, collaborate with other scientists and engineers, and be part of Amazon’s vibrant and diverse global science community. * Publish your innovation in top-tier academic venues and hone your presentation skills. * Be inspired by challenges and opportunities to invent cutting-edge techniques in your area(s) of expertise.