How dynamic lookahead improves speech recognition

Determining on the fly how much additional audio to process to resolve ambiguities increases accuracy while reducing latency relative to fixed-lookahead approaches.

Automatic speech recognition (ASR) models, which convert speech into text, come in two varieties, causal and noncausal. A causal model processes speech as it comes in; to determine the correct interpretation of the current frame (discrete chunk) of audio, it can use only the frames that preceded it. A noncausal model waits until an utterance is complete; in interpreting the current frame, it can use both the frames that preceded it and those that follow it.

Causal models tend to have lower latencies, since they don’t have to wait for frames to come in, but noncausal models tend to be more accurate, because they have additional contextual information. Many ASR models try to strike a balance between the two approaches by using lookahead: they let a few additional frames come in before deciding on the interpretation of the current frame. Sometimes, however, those additional frames don’t include the crucial bit of information that could resolve a question of interpretation, and sometimes, the model would have been just as accurate without them.

In a paper we presented at this year’s International Conference on Machine Learning (ICML), we describe an ASR model that dynamically determines lookahead for each frame, based on the input.

We compared our model to a causal model and two standard types of lookahead models and found that, across the board, our model achieved lower error rates than any of the baselines. At the same time, for a given error rate, it achieved lower latencies than either of the earlier lookahead models.

Computational graph

We represent the computations executed by our model with a computational graph. From left to right, the graph depicts successive time steps in the processing of input frames; from bottom to top, it depicts successive layers of the ASR network, from input to output. Edges in the graph depict causal relationships between nodes at past time steps and nodes at the current time step, and they also depict dependency relationships between nodes at future time steps and the current output.

Computational graph.png
In this computational graph, grey arrows depict causal relationships between network nodes at different time steps, and blue arrows depict relationships between nodes at future time steps and the current output node (vi3).
Lookahead model.png
The adjacency matrix of a standard lookahead model.
Dynamic lookahead model.png
A mask generated dynamically by our model.

We represent each layer in the graph, in turn, with an adjacency matrix, which maps all the layer’s nodes against those from the prior layer; the value in any cell of the matrix indicates the row node’s dependency on the column node.

The matrix of a purely causal model is divided by a diagonal from top left to bottom right; all the values to the right of the diagonal are zero, because there are no dependencies between future time steps and the current time step. An entirely noncausal model, by contrast, has a full matrix. A standard lookahead model has a diagonal that’s offset by as many frames as it looks ahead.

Our goal is to train a scheduler that generates adjacency matrices on the fly, with differing degrees of lookahead for different rows of the matrices. We call these matrices masks, because they mask out parts of the adjacency matrix. 


Ultimately, we want the values of the masks to be binary: either we look ahead to a future frame or we don’t. But the loss function we use during training must be differentiable, so we can use the standard gradient descent algorithm to update the model weights. Consequently, during training, we allow fractional values in the adjacency matrices.

In a process known as annealing, over the course of successive training epochs, we force the values of the adjacency matrix to diverge more and more, toward either 1 or 0. At inference time, the values output by the model will still be fractional, but they will be close enough to 1 or 0 that we can produce the adjacency matrix by simple rounding.

Dependency weights across successive time steps for several different cells in an adjacency matrix as the annealing “temperature” is turned up during training. Fractional values gradually resolve to 1 or 0.


A lookahead ASR model needs to balance accuracy and latency, and with our architecture, we strike that balance through the choice of loss function during training.

A naïve approach would be simply to have two terms in the loss function, one that penalizes error and one that penalizes total lookahead within the masks as a proxy for latency. But we take a more sophisticated approach.

During training, for every computational graph generated by our model, we compute the algorithmic latency for each output. Recall that, during training, the values in the graph can be fractional; we define algorithmic latency as the number of time steps between the current output node and the future input node whose dependency path to the current node has the highest weight.

Lookahead compute graph.png
In this example, the algorithmic latency for node v3i is two frames, since it depends on the input value v0i+2.

This allows us to compute the average algorithmic latency for all the examples in our training set and, consequently, to regularize the latency measure we use during training. That is, the latency penalty is not absolute but relative to the average lookahead necessary to ensure model accuracy.

In a separate set of experiments, we used a different notion of latency: computational latency, rather than algorithmic latency. There, the key was to calculate how much of its backlogged computations the model could get through in each time step; the unfinished computations after the final time step determined the user-perceived latency.

Computational latency.png
With computational (rather than algorithmic) latency, the computational backlog after the final time step determines the user-perceived latency (UPL).

 As with any multi-objective loss function, we can tune the relative contribution of each loss term. Below are masks generated by two versions of our model for the same input data. Both versions were trained using algorithmic latency, but in one case (right), the latency penalty was more severe than in the other. As can be seen, the result is a significant drop in latency, but at the cost of an increase in error rate.

Different latency penalties.png
Masks generated by two different models, one (left) trained with a lower latency penalty and one (right) with a higher penalty.

We compared our model’s performance to four baselines: one was a causal model, with no lookahead; one was a layerwise model, which used the same lookahead for each frame; one was a chunked model, which executes a lookahead once, catches up with it, then executes another lookahead; and the last was a version of our dynamic-lookahead model, except using the standard latency penalty term. We also tested two versions of our model, one built with the Conformer architecture and one with the Transformer.

For the fixed-lookahead baselines, we considered three different lookahead intervals: two frames, five frames, and 10 frames. Across the board, our models were more accurate than all four baselines, while also achieving lower latencies.

Dynamic-lookahead results.png
Results of dynamic-lookahead experiments.

Acknowledgments: Martin Radfar, Ariya Rastrow, Athanasios Mouchtaris

Related content

US, WA, Seattle
The Amazon Economics Team is hiring Economist Interns. We are looking for detail-oriented, organized, and responsible individuals who are eager to learn how to work with large and complicated data sets to solve real-world business problems. 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 future job market 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 We are open to hiring candidates to work out of one of the following locations: Seattle, WA, USA
GB, Cambridge
Our team undertakes research together with multiple organizations to advance the state-of-the-art in speech technologies. We not only work on giving Alexa, the ground-breaking service that powers Echo, her voice, but we also develop cutting-edge technologies with Amazon Studios, the provider of original content for Prime Video. Do you want to be part of the team developing the latest technology that impacts the customer experience of ground-breaking products? Then come join us and make history. We are looking for a passionate, talented, and inventive Senior Applied Scientist with a background in Machine Learning to help build industry-leading Speech, Language and Video technology. As a Senior Applied Scientist at Amazon you will work with talented peers to develop novel algorithms and modelling techniques to drive the state of the art in speech and vocal arts synthesis. Position Responsibilities: - Participate in the design, development, evaluation, deployment and updating of data-driven models for digital vocal arts applications. - Participate in research activities including the application and evaluation and digital vocal and video arts techniques for novel applications. - Research and implement novel ML and statistical approaches to add value to the business. - Mentor junior engineers and scientists. We are open to hiring candidates to work out of one of the following locations: Cambridge, GBR
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 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 reduced-form causal analysis and/or structural economic modeling methods to evaluate the impact of policies on employee outcomes, and examine how external labor market and economic conditions impact Amazon's ability to hire and retain talent. 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. We are open to hiring candidates to work out of one of the following locations: Arlington, VA, USA
US, WA, Seattle
We are expanding our Global Risk Management & Claims team and insurance program support for Amazon’s growing risk portfolio. This role will partner with our risk managers to develop pricing models, determine rate adequacy, build underwriting and claims dashboards, estimate reserves, and provide other analytical support for financially prudent decision making. As a member of the Global Risk Management team, this role will provide actuarial support for Amazon’s worldwide operation. Key job responsibilities ● Collaborate with risk management and claims team to identify insurance gaps, propose solutions, and measure impacts insurance brings to the business ● Develop pricing mechanisms for new and existing insurance programs utilizing actuarial skills and training in innovative ways ● Build actuarial forecasts and analyses for businesses under rapid growth, including trend studies, loss distribution analysis, ILF development, and industry benchmarks ● Design actual vs expected and other metrics dashboards to assist decision makings in pricing analysis ● Create processes to monitor loss cost and trends ● Propose and implement loss prevention initiatives with impact on insurance pricing in mind ● Advise underwriting decisions with analysis on driver risk profile ● Support insurance cost budgeting activities ● Collaborate with external vendors and other internal analytics teams to extract insurance insight ● Conduct other ad hoc pricing analyses and risk modeling as needed We are open to hiring candidates to work out of one of the following locations: Arlington, VA, USA | New York, NY, USA | Seattle, WA, USA
US, NY, New York
The Amazon SCOT Forecasting team seeks a Senior Applied Scientist to join our team. Our research team conducts research into the theory and application of reinforcement learning. This research is shared in top journals and conferences and has a significant impact on the field. Through our launch of several Deep RL models into production, our work also affects decision making in the real world. Members of our group have varied interests—from the mathematical foundations of reinforcement learning, to language modeling, to maintaining the performance of generative models in the face of copyrights, and more. Recent work has focused on sample efficiency of RL algorithms, treatment effect estimation, and RL agents integrating real-world constraints, as applied in supply chains. Previous publications include: - Linear Reinforcement Learning with Ball Structure Action Space - Meta-Analysis of Randomized Experiments with Applications to Heavy-Tailed Response Data - A Few Expert Queries Suffices for Sample-Efficient RL with Resets and Linear Value Approximation - Deep Inventory Management - What are the Statistical Limits of Offline RL with Linear Function Approximation? Working collaboratively with a group of fellow scientists and engineers, you will identify complex problems and develop solutions in the RL space. We encourage collaboration across teammates and their areas of specialty, leading to creative and ambitious projects with the goal of publication and production. Key job responsibilities - Drive collaborative research and creative problem solving - Constructively critique peer research; mentor junior scientists - Create experiments and prototype implementations of new algorithms and techniques - Collaborate with engineering teams to design and implement software built on these new algorithms - Contribute to progress of the Amazon and broader research communities by producing publications We are open to hiring candidates to work out of one of the following locations: New York, NY, USA
US, CA, Virtual Location - California
If you are interested in this position, please apply on Twitch's Career site About Us: Launched in 2011, Twitch is a global community that comes together each day to create multiplayer entertainment: unique, live, unpredictable experiences created by the interactions of millions. We bring the joy of co-op to everything, from casual gaming to world-class esports to anime marathons, music, and art streams. Twitch also hosts TwitchCon, where we bring everyone together to celebrate and grow their personal interests and passions. We're always live at Twitch. About the Role: As a Data Scientist, Analytics member of the Data Platform - Insights team, you'll provide data analysis and support for platform, service, and operational engineering teams at Twitch, shaping the way success is measured. Defining what questions should be asked and scaling analytics methods and tools to support our growing business. Additionally, you will help support the vision for business analytics, solutions architecture for data related business constructs, as well as tactical execution such as experiment analysis and campaign performance reporting. You are paving the way for high-quality, high-velocity decisions and will report to the Manager, Data Science. For this role, we're looking for an experienced data staff who will oversee data instrumentation, dashboard/report building, metrics reviews, inform team investments, guidance on success/failure metrics and ad-hoc analysis. You will also work with technical and non-technical staff members throughout the company, and your effort will have an impact on hundreds of partners at Twitch You Will: - Work with members of Platforms & Services to guide them towards better decision making from the available data. - Promote data knowledge and insights through managing communications with partners and other teams, collaborate with colleagues to complete data projects and ensure all parties can use the insights to further improve. - Maintain a customer-centric focus while being a domain and product expert through data, develop trust amongst peers, and ensure that the teams and programs have access to data to make decisions - Manage ambiguous problems and adapt tools to answer complicated questions. - Identify the trade-offs between speed and quality of different approaches. - Create analytical frameworks to measure team success by partnering with teams to establish success metrics, create approaches to track the data and troubleshoot errors, measure and evaluate the data to develop a common language for all colleagues to understand these metrics. - Operationalize data processes to provide partners with ad-hoc analysis, automated dashboards, and self-service reporting tools so that everyone gets a good sense of the state of the business Perks: - Medical, Dental, Vision & Disability Insurance - 401(k), Maternity & Parental Leave - Flexible PTO - Commuter Benefits - Amazon Employee Discount - Monthly Contribution & Discounts for Wellness Related Activities & Programs (e.g., gym memberships, off-site massages), -Breakfast, Lunch & Dinner Served Daily - Free Snacks & Beverages We are open to hiring candidates to work out of one of the following locations: Irvine, CA, USA | Seattle, WA, USA | Virtual Location - CA
US, WA, Bellevue
Have you ever ordered a product on Amazon and when that box with the smile arrived you wondered how it got to you so fast? Have you wondered where it came from and how much it cost Amazon to deliver it to you? Have you also wondered what are different ways that the transportation assets can be used to delight the customer even more. If so, the Amazon transportation Services, Product and Science is for you . We manage the delivery of tens of millions of products every week to Amazon’s customers, achieving on-time delivery in a cost-effective manner. We are looking for an enthusiastic, customer obsessed Applied Scientist with strong scientific thinking, good software and statistics experience, skills to help manage projects and operations, improve metrics, and develop scalable processes and tools. The primary role of an Applied Scientist within Amazon is to address business challenges through building a compelling case, and using data to influence change across the organization. This individual will be given responsibility on their first day to own those business challenges and the autonomy to think strategically and make data driven decisions. Decisions and tools made in this role will have significant impact to the customer experience, as it will have a major impact on how we operate the middle mile network. Ideal candidates will be a high potential, strategic and analytic graduate with a PhD in (Operations Research, Statistics, Engineering, and Supply Chain) ready for challenging opportunities in the core of our world class operations space. Great candidates have a history of operations research, machine learning , and the ability to use data and research to make changes. This role requires robust skills in research and implementation of scalable products and models . This individual will need to be able to work with a team, but also be comfortable making decisions independently, in what is often times an ambiguous environment. Responsibilities may include: - Develop input and assumptions based preexisting models to estimate the costs and savings opportunities associated with varying levels of network growth and operations - Creating metrics to measure business performance, identify root causes and trends, and prescribe action plans - Managing multiple projects simultaneously - Working with technology teams and product managers to develop new tools and systems to support the growth of the business - Communicating with and supporting various internal stakeholders and external audiences We are open to hiring candidates to work out of one of the following locations: Bellevue, WA, USA
US, CA, Los Angeles
The Alexa team is looking for a passionate, talented, and inventive Applied Scientist with a strong machine learning background, to help build industry-leading Speech and Language technology. Key job responsibilities As an Applied Scientist with the Alexa team, you will work with talented peers to develop novel algorithms and modeling techniques to advance the state of the art in spoken language understanding. Your work will directly impact our customers in the form of products and services that make use of speech and language technology. You will leverage Amazon’s heterogeneous data sources and large-scale computing resources to accelerate advances in spoken language understanding. About the team The Alexa team has a mission to push the envelope in Automatic Speech Recognition (ASR), Natural Language Understanding (NLU), and Audio Signal Processing, in order to provide the best-possible experience for our customers. We are open to hiring candidates to work out of one of the following locations: Los Angeles, CA, USA
US, WA, Seattle
Are you fascinated by the power of Natural Language Processing (NLP) and Large Language Models (LLM) to transform the way we interact with technology? Are you passionate about applying advanced machine learning techniques to solve complex challenges in the e-commerce space? If so, Amazon's International Seller Services team has an exciting opportunity for you as an Applied Scientist. At Amazon, we strive to be Earth's most customer-centric company, where customers can find and discover anything they want to buy online. Our International Seller Services team plays a pivotal role in expanding the reach of our marketplace to sellers worldwide, ensuring customers have access to a vast selection of products. As an Applied Scientist, you will join a talented and collaborative team that is dedicated to driving innovation and delivering exceptional experiences for our customers and sellers. You will be part of a global team that is focused on acquiring new merchants from around the world to sell on Amazon’s global marketplaces around the world. The position is based in Seattle but will interact with global leaders and teams in Europe, Japan, China, Australia, and other regions. Join us at the Central Science Team of Amazon's International Seller Services and become part of a global team that is redefining the future of e-commerce. With access to vast amounts of data, cutting-edge technology, and a diverse community of talented individuals, you will have the opportunity to make a meaningful impact on the way sellers engage with our platform and customers worldwide. Together, we will drive innovation, solve complex problems, and shape the future of e-commerce. Please visit for more information Key job responsibilities - Apply your expertise in LLM models to design, develop, and implement scalable machine learning solutions that address complex language-related challenges in the international seller services domain. - Collaborate with cross-functional teams, including software engineers, data 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 seller performance and customer experiences across various international marketplaces. - Continuously explore and evaluate state-of-the-art NLP techniques and methodologies to improve the accuracy and efficiency of language-related 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. We are open to hiring candidates to work out of one of the following locations: Seattle, WA, USA
US, CA, Palo Alto
We’re working to improve shopping on Amazon using the conversational capabilities of large language models. We are open to hiring candidates to work out of one of the following locations: Palo Alto, CA, USA