Alexa’s text-to-speech research at Interspeech 2022

Highlighted papers focus on transference — of prosody, accent, and speaker identity.

Interspeech, the world’s largest and most comprehensive conference on the science and technology of spoken-language processing, took place last week in Incheon, Korea, with Amazon as a platinum sponsor. Amazon Science asked three of Alexa AI’s leading scientists — in the fields of speech, spoken-language-understanding, and text-to-speech — to highlight some of Amazon’s contributions to the conference.

In this installment, Antonio Bonafonte, a senior applied scientist in the Amazon Text-to-Speech group, highlights work on transference — of prosody, accent, and speaker identity — in text-to-speech.

This year, the Amazon Text-to-Speech organization presented more than a dozen papers at Interspeech 2022. Amazon TTS gives Alexa its voice, working every day to add more expressiveness and conversational awareness. Here we highlight some of papers that illustrate what we are doing in those directions.

Expressive and contextually appropriate prosody

Neural text-to-speech (TTS) techniques have made the speech produced by TTS systems much more natural. To make the prosody of the speech more expressive and context appropriate as well, researchers have done considerable work on learning prosody representations from ground-truth speech.

The paper “CopyCat2: A single model for multi-speaker TTS and many-to-many fine-grained prosody transfer”, by Sri Karlapati and coauthors, proposes a model that learns word-level speaker-independent prosody representations from multispeaker speech. These representations can be used for fine-grained prosody transfer from multiple source speakers to multiple target speakers. Furthermore, predicting the word-level prosody representations from text results in a TTS model with improved naturalness and appropriateness.

CopyCat2.png
The CopyCat2 architecture.

The word-level prosodic representation is split into two components, one for timing and rhythm and a second for other prosodic characteristics. The figure above shows how the second component is learned using a conditional variational autoencoder. The input mel-spectrogram (X), which represents the speech signal as energies in certain frequency bands, is compressed into a sequence of vectors (Z), one per word. Those vectors are then used to reconstruct the mel-spectrogram.

Related content
New voice for Alexa’s Reading Sidekick feature avoids the instabilities common to models with variable prosody.

The decoder is conditioned on the phonemes and the speaker, so it captures speaker-independent prosody information, and a similar approach is used to learn speaker-independent word-level representations of timing aspects.

To use CopyCat2 as a text-to-speech model, the researchers train an additional model to predict the parameters of the prosodic-word-embedding distribution (Z) from BERT embeddings. In tests involving a multispeaker US English dataset of varied styles, including news, facts, and greetings, they compared their approach to a strong TTS baseline with contextually appropriate prosody and copy-synthesized speech. They found that their model reduced the gap in naturalness between synthetic and real speech by 22.79%.

Reducing the data required to build expressive voices

Training a state-of-the-art TTS model is usually a data-intensive process, and building a portfolio of voices in multiple styles and languages compounds the data requirement.

In the paper “Low-data? No problem: low-resource, language-agnostic conversational text-to-speech via F0-conditioned data augmentation”, Giulia Comini et al. propose a methodology to build expressive text-to-speech voices using only one hour of expressive speech from the target speaker. The method requires 8–10 hours of neutral speech — that is, speech with a limited range of expression — from another speaker, a significant reduction from previous methods.

Low data.png
A new approach to building expressive text-to-speech voices can make do with only an hour of expressive speech from the target speaker.

The authors propose to convert the neutral data from the supporting speaker to the target-speaker identity, while maintaining the target speaker’s expressive style. They use a modification of the original CopyCat prosody transfer model. As shown in the figure, the CopyCat parallel decoder regenerates the mel-spectrogram from the speaker embedding; the fundamental frequency (F0), or perceived pitch of individual phonemes; the phonetic representation; and the output of the CopyCat reference encoder. The reference encoder captures the information from the source mel-spectrogram that is not explicitly given to the decoder, — i.e., phonemes, with their duration and F0, and the speaker embedding.

Related content
Users find speech with transferred expression 9% more natural than standard synthesized speech.

The model is trained with the expressive speech of the target speaker and neutral speech from the supporting speaker. Once the model is trained, the mel-spectrogram of the supporting data is transformed into augmented expressive data for the target speaker. The CopyCat decoder is conditioned on the target speaker embedding and on an expressive F0 contour generated from the text and the speaker embedding by an independent model trained with the same data.

The paper shows that the F0 distribution of the augmented data resembles that of the target speaker. They also show that their data augmentation approach improves on one that does not use F0 conditioning.

Alexa multilingual models

Amazon has developed a shared neural TTS model for several speakers and languages that can extend a synthetic voice trained on data in only one language into other languages. For instance, the technology allows the English-language Alexa feminine-sounding voice to speak fluent Spanish in US multilingual homes. Similarly, Alexa’s English-language US masculine-sounding voice already has a British accent in the UK and speaks Spanish in the US, French in Canada, and German in Germany.

Related content
Neural text-to-speech enables new multilingual model to use the same voice for Spanish and English responses.

Alexa communicates on a wide variety of topics, and the style of speech should match the textual content. Transferring styles across languages while maintaining a fixed speaker identity, however, is challenging.

In the paper “Cross-lingual style transfer with conditional Prior VAE and style loss”, Dino Ratcliffe et al. propose an architecture for cross-lingual style transfer. Specifically, they improve the Spanish representation across four styles — newscaster, DJ, excited, and disappointed — while maintaining a single speaker identity for which only English samples are available.

Cross-lingual style transfer.png
A new approach to cross-lingual style transfer groups utterances of the same style together irrespective of language.

They achieve this by using a learned-conditional-prior variational autoencoder (LCPVAE), a hierarchical variational-autoencoder (VAE) approach.

The approach introduces a secondary VAE, which is conditioned on one-hot-encoded style information; that is, the style code has as many bits as there are styles, and a 1 at exactly one spot denotes a particular style. This results in a structured embedding space, which groups together utterances of the same style irrespective of language.

Related content
Papers focus on speech conversion and data augmentation — and sometimes both at once.

As can be seen in the figure, the TTS decoder generates the mel-spectrogram from the speaker embedding, language, phonemes, and the style embedding. During training, the style embeddings are generated by the LCPVAE using the one-hot code and the reference mel-spectrogram; at inference, the style embedding is the centroid of the embeddings for a particular style. The model’s loss function includes a style classification term that steers the generated mel-spectrogram toward the same style as the reference spectrogram.

Based on subjective evaluations (MUSHRA), this approach shows significant improvements on cross-lingual style representation in all four styles, DJ (2.8%), excited (5.3%), disappointed (3.5%) and newscaster (2.3%), without compromising speaker similarity and in-lingual style representation.

Creating new characters

Current TTS technology can produce realistic synthetic speech for sample voice identities seen during training. But speech synthesis with speakers unseen during training, without post-training adaptation, remains a big challenge. Synthesis with a new voice often means creating high-quality data to train a generative model.

Related content
Thanks to a set of simple abstractions, models with different architectures can be integrated and optimized for particular hardware accelerators.

Normalizing flows are generative models with tractable distributions, where sampling and density evaluation can be both exact and efficient. In “Creating new voices using normalizing flows”, Piotr Biliński and his colleagues investigate the ability of normalizing flows in TTS and voice conversion modes to extrapolate from speakers observed during training to unseen speaker identities — without any recordings of those speakers, and therefore without the possibility of target speaker adaptation.

Their approach is based on the Flow-TTS model, but instead of using it to generate synthetic speech of seen speakers, they adapted it to create new voices. Key contributions include adding the ability to sample new speakers, introducing voice conversion mode, and comparing it to TTS mode.

Normalizing flows.png
Instead of using normalizing flows to synthesize the speech of seen speakers, Amazon researchers adapted them to create new voices.

The architecture of the model consists of an invertible transformation based on normalizing flows. The design allows for lossless reconstruction of a mel-spectrogram from a representational space (z) given conditions (θ) such as speaker embedding. In text-to-speech mode, sampling z from the prior distribution and running the inverse transformation allows us to generate the mel-spectrogram given the conditions θ.

To apply the model in voice conversion mode, we map the source mel-spectrogram to a latent representation z using as condition the source-speaker embedding. Then, the latent representation z is converted back to a mel-spectrogram using the speaker embedding of the target speaker. To generate speaker embeddings of new voices, we train a separate neural network that generates plausible speaker embeddings for a given regional English variant.

Extensive evaluations demonstrate that the proposed approach systematically obtains state-of-the-art performance in zero-shot speech synthesis and allows us to create voices distinct from those in the training set. In addition, the authors find that as the level of conditioning to the model is increased, voice conversion and TTS modes can be used interchangeably.

Related content

US, CA, Sunnyvale
At Amazon Fashion, we are obsessed with making Amazon Fashion the most loved fashion destinations globally. We're searching for Computer Vision pioneers who are passionate about technology, innovation, and customer experience, and who are enthusiastic about making a lasting impact on the industry. You'll be working with talented scientists, engineers, and product managers to innovate on behalf of our customers. If you're fired up about being part of a dynamic, driven team, then this is your moment to join us on this exciting journey and change the world of eCommerce forever Key job responsibilities As a Applied Scientist, you will be at the forefront to define, own and drive the science that span multiple machine learning models and enabling multiple product/engineering teams and organizations. You will partner with product management and technical leadership to identify opportunities to innovate customer facing experiences. You will identify new areas of investment and work to align product roadmaps to deliver on these opportunities. As a science leader, you will not only develop unique scientific solutions, but more importantly influence strategy and outcomes across different Amazon organizations such as Search, Personalization and more. This role is inherently cross-functional and requires a strong ability to communicate, influence and earn the trust of software engineers, technical and business leadership. We are open to hiring candidates to work out of one of the following locations: Sunnyvale, CA, 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, 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 econ-internship@amazon.com. We are open to hiring candidates to work out of one of the following locations: Seattle, WA, USA
US, NY, New York
Amazon is investing heavily in building a world-class advertising business, and we are responsible for defining and delivering a collection of self-service performance advertising products that drive discovery and sales. We deliver billions of ad impressions and millions of clicks daily and break fresh ground to create world-class products. We are highly motivated, collaborative, and fun-loving with an entrepreneurial spirit and bias for action. With a broad mandate to experiment and innovate, we are growing at an unprecedented rate with a seemingly endless range of new opportunities. Our systems and algorithms operate on one of the world's largest product catalogs, matching shoppers with advertised products with a high relevance bar and strict latency constraints. Sponsored Products Detail Page Blended Widgets team is chartered with building novel product recommendation experiences. We push the innovation frontiers for our hundreds of millions of customers WW to aid product discovery while helping shoppers to find relevant products easily. Our team is building differentiated recommendations that highlight specific characteristics of products (either direct attributes, inferred or machine learned), and leveraging generative AI to provide interactive shopping experiences. We are looking for a Senior Applied Scientist who can delight our customers by continually learning and inventing. Our ideal candidate is an experienced Applied Scientist who has a track-record of performing deep analysis and is passionate about applying advanced ML and statistical techniques to solve real-world, ambiguous and complex challenges to optimize and improve the product performance, and who is motivated to achieve results in a fast-paced environment. The position offers an exceptional opportunity to grow your technical and non-technical skills and make a real difference to the Amazon Advertising business. As a Senior Applied Scientist on this team, you will: * Be the technical leader in Machine Learning; lead efforts within this team and collaborate across teams * Rapidly design, prototype and test many possible hypotheses in a high-ambiguity environment, perform hands-on analysis and modeling of enormous data sets to develop insights that improve shopper experiences and merchandise sales * Drive end-to-end Machine Learning projects that have a high degree of ambiguity, scale, complexity. * Build machine learning models, perform proof-of-concept, experiment, optimize, and deploy your models into production; work closely with software engineers to assist in productionizing your ML models. * Establish scalable, efficient, automated processes for large-scale data analysis, machine-learning model development, model validation and serving. * Research new and innovative machine learning approaches. * Promote the culture of experimentation and applied science at Amazon Team video https://youtu.be/zD_6Lzw8raE We are also open to consider the candidate in Seattle, or Palo Alto. We are open to hiring candidates to work out of one of the following locations: New York, NY, USA
US, VA, Arlington
Amazon Advertising is one of Amazon's fastest growing and most profitable businesses, responsible for defining and delivering a collection of advertising products that drive discovery and sales. As a core product offering within our advertising portfolio, Sponsored Products (SP) helps merchants, retail vendors, and brand owners succeed via native advertising, which grows incremental sales of their products sold through Amazon. The SP team's primary goals are to help shoppers discover new products they love, be the most efficient way for advertisers to meet their business objectives, and build a sustainable business that continuously innovates on behalf of customers. Our products and solutions are strategically important to enable our Retail and Marketplace businesses to drive long-term growth. We deliver billions of ad impressions and millions of clicks and break fresh ground in product and technical innovations every day! The Search Sourcing and Relevance team parses billions of ads to surface the best ad to show to Amazon shoppers. The team strives to understand customer intent and identify relevant ads that enable them to discover new and alternate products. This also enables sellers on Amazon to showcase their products to customers, which may, at times, be buried deeper in the search results. By showing the right ads to customers at the right time, this team improves the shopper experience, increase advertiser ROI, and improves long-term monetization. This is a talented team of machine learning scientists and software engineers working on complex solutions to understand the customer intent and present them with ads that are not only relevant to their actual shopping experience but also non-obtrusive. This area is of strategic importance to Amazon Retail and Marketplace business, driving long term growth. Key job responsibilities As a Senior Applied Scientist on this team, you will: - Be the technical leader in Machine Learning; lead efforts within this team and across other teams. - Perform hands-on analysis and modeling of enormous data sets to develop insights that increase traffic monetization and merchandise sales, without compromising the shopper experience. - Drive end-to-end Machine Learning projects that have a high degree of ambiguity, scale, complexity. - Build machine learning models, perform proof-of-concept, experiment, optimize, and deploy your models into production; work closely with software engineers to assist in productionizing your ML models. - Run A/B experiments, gather data, and perform statistical analysis. - Establish scalable, efficient, automated processes for large-scale data analysis, machine-learning model development, model validation and serving. - Research new and innovative machine learning approaches. - Recruit Applied Scientists to the team and provide mentorship. About the team Amazon is investing heavily in building a world-class advertising business. This team defines and delivers a collection of advertising products that drive discovery and sales. Our solutions generate billions in revenue and drive long-term growth for Amazon’s Retail and Marketplace businesses. We deliver billions of ad impressions, millions of clicks daily, and break fresh ground to create world-class products. We are a highly motivated, collaborative, and fun-loving team with an entrepreneurial spirit - with a broad mandate to experiment and innovate. You will invent new experiences and influence customer-facing shopping experiences to help suppliers grow their retail business and the auction dynamics that leverage native advertising; this is your opportunity to work within the fastest-growing businesses across all of Amazon! Define a long-term science vision for our advertising business, driven from our customers' needs, translating that direction into specific plans for research and applied scientists, as well as engineering and product teams. This role combines science leadership, organizational ability, technical strength, product focus, and business understanding. We are open to hiring candidates to work out of one of the following locations: Arlington, VA, USA
US, WA, Seattle
Amazon Advertising Impact Team is looking for a Senior Economist to help translate cutting-edge causal inference and machine learning research into production solutions. The individual will have the opportunity to shape the technical and strategic vision of a highly ambiguous problem space, and deliver measurable business impacts via cross-team and cross-functional collaboration. Amazon is investing heavily in building a world class advertising business. Our advertising products are strategically important to Amazon’s Retail and Marketplace businesses for driving long-term growth. The mission of the Advertising Impact Team is to make our advertising products the most customer-centric in the world. We specialize in measuring and modeling the short- and long-term customer behavior in relation to advertising, using state of the art econometrics and machine learning techniques. With a broad mandate to experiment and innovate, we are constantly advancing our experimentation methodology and infrastructure to accelerate learning and scale impacts. We are highly motivated, collaborative and fun-loving with an entrepreneurial spirit and bias for action. Key job responsibilities • Function as a technical leader to shape the strategic vision and the science roadmap of a highly ambiguous problem space • Develop economic theory and deliver econometrics and machine learning models to optimize advertising strategies on behalf of our customers • Design, execute, and analyze experiments to verify the efficacy of different scientific solutions in production • Partner with cross-team technical contributors (scientists, software engineers, product managers) to implement the solution in production • Write effective business narratives and scientific papers to communicate to both business and technical audience, including the most senior leaders of the company We are open to hiring candidates to work out of one of the following locations: New York, NY, USA | Seattle, WA, 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, 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
Interested in using the latest, cutting edge machine learning and science to improve the Amazon employee experience? This role provides applied science leadership to the organization that develops and delivers data-driven insights, personalization, and nudges into Amazon's suite of talent management products to help managers, employees, and organizational leaders make better decisions and have better, more equitable outcomes. Key job responsibilities As the Principal Applied Scientist for GTMC SIERRA, you will be responsible for providing scientific thought leadership over multiple applied science and engineering teams. Each of these teams has rapidly evolving and complex demands to define, develop, and deliver scalable products that make work easier, more efficient, and more rewarding for Amazonians. These are some of Amazon’s most strategic technical investments in the people space and support Amazon’s efforts to be Earth’s Best Employer. In this role you will have a significant impact on 1.5 million Amazonians and the communities Amazon serves. You will also play a critical role in the organization's business planning, work closely with senior executives to develop goals and resource requirements, influence our long-term technical and business strategy, and help hire and develop engineering and science talent. You will provide science thought leadership and support to business partners, helping them use the best scientific methods and science-driven tools to solve current and upcoming challenges and deliver efficiency gains in a changing market. About the team Global Talent Management & Compensation (GTMC) SIERRA (Science, Insights, Experience, Research, Reporting & Analytics) is a horizontal, multi-disciplinary organization whose mission is to be a force multiplier for the broader GTMC organization and our key customer cohorts. We accomplish this by using our expertise in data analytics and science, economics, machine learning (ML), UX, I/O psychology, and engineering to build insights and experiences that raise the bar in understanding and shaping decision making at scale by integrating within and across talent journeys as well as through self-service tools and closed loop mechanisms outside of those journeys. Our portfolio of products spans foundational data sources, metrics, and research through to finished features and products that our end-customers interact with on a daily basis. We are open to hiring candidates to work out of one of the following locations: Seattle, WA, USA
US, WA, Seattle
The economics team within Recruiting Engine uses economics, behavioral science, statistics, and machine learning to proactively identify mechanisms and process improvements which simultaneously improve Amazon and the lives, well-being, and the value of work to Amazonians. We are an interdisciplinary team, which uses a range of approaches to develop and deliver solutions that measurably achieve this goal. We are looking for an Economist who is able to provide structure around complex business problems, hone those complex problems into specific, scientific questions, and test those questions to generate insights. The ideal candidate will work with various science, engineering, operations and analytics teams 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. She/He/They will produce robust, objective research results and insights which can be communicated to a broad audience inside and outside of Amazon. Ideal candidates will work closely with business partners to develop science that solves the most important business challenges. She/He/They will work well in a team setting with individuals from diverse disciplines and backgrounds. She/He/They will serve as an ambassador for science and a scientific resource for business teams. Ideal candidates will own the development of scientific models and manage the data analysis, modeling, and experimentation that is necessary for estimating and validating the model. They will be customer-centric – clearly communicating scientific approaches and findings to business leaders, listening to and incorporate their feedback, and delivering successful scientific solutions. We are open to hiring candidates to work out of one of the following locations: Arlington, VA, USA | Seattle, WA, USA