Search results

Although deep neural networks have enabled accurate large-vocabulary speech recognition, training them requires thousands of hours of transcribed data, which is time-consuming and expensive to collect. So Amazon scientists have been investigating techniques that will let Alexa learn with minimal human involvement, techniques that fall in the categories of unsupervised and semi-supervised learning.

In experiments involving sound recognition, technique reduces error rate by 15% to 30%.

The 2018 Alexa Prize featured eight student teams from four countries, each of which adopted distinctive approaches to some of the central technical questions in conversational AI. We survey those approaches in a paper we released late last year, and the teams themselves go into even greater detail in the papers they submitted to the latest Alexa Prize Proceedings. Here, we touch on just a few of the teams’ innovations.

To ensure that Alexa Prize contestants can concentrate on dialogue systems — the core technology of socialbots — Amazon scientists and engineers built a set of machine learning modules that handle fundamental conversational tasks and a development environment that lets contestants easily mix and match existing modules with those of their own design.

This Sunday's Super Bowl between the New England Patriots and the Los Angeles Rams is expected to draw more than 100 million viewers, some of whom will have Alexa-enabled devices within range of their TV speakers. When Amazon's new Alexa ad airs, and Forest Whitaker asks his Alexa-enabled electric toothbrush to play his podcast, how will we prevent viewers’ devices from mistakenly waking up?

Many of today’s most popular AI systems are, at their core, classifiers. They classify inputs into different categories: this image is a picture of a dog, not a cat; this audio signal is an instance of the word “Boston”, not the word “Seattle”; this sentence is a request to play a video, not a song. But what happens if you need to add a new class to your classifier — if, say, someone releases a new type of automated household appliance that your smart-home system needs to be able to control?

Each with our own experiences and unique value, we plan to tackle the challenge of creating an active listening, heartwarming, and understanding social bot.

We propose to develop solutions to these challenges to allow our socialbot, Tartan, to approach a human capacity for natural conversations.

We envision to build an enjoyable, personalized, and dynamic social-bot that is able to engage in deep conversations across topics and adapt to different user types.

Our team envisions creating a socialbot that will be able to engage in memorable interactions, and can tailor the conversation to reflect the interests of each unique individual.

We aim to have a holistic approach towards achieving a multi-turn, on-topic and engaging conversation by designing our systems based on the principles of mixed-initiative.

Our socialbot will fill the gap between production-ready agents and state-of-the-art NLP research.

Our scientific approach focuses on developing novel models for neural natural language generation and dialogue management.

We love academic research in artificial intelligence as well as cooperating on practical projects with companies.

We expect our socialbot, Emora, to be your companion who will care about you, learn from you, share thoughts and feelings with you, and most importantly, always be there for you.

Machine learning systems often act on “features” extracted from input data. In a natural-language-understanding system, for instance, the features might include words’ parts of speech, as assessed by an automatic syntactic parser, or whether a sentence is in the active or passive voice.

The team envisions a future where personal, sensible social conversational agents focus on conversational diversity and consistency in personalized dialogue.

Developing a new natural-language-understanding system usually requires training it on thousands of sample utterances, which can be costly and time-consuming to collect and annotate. That’s particularly burdensome for small developers, like many who have contributed to the library of more than 70,000 third-party skills now available for Alexa.

Neural networks have been responsible for most of the top-performing AI systems of the past decade, but they tend to be big, which means they tend to be slow. That’s a problem for systems like Alexa, which depend on neural networks to process spoken requests in real time.

In May 2018, Amazon launched Alexa’s Remember This feature, which enables customers to store “memories” (“Alexa, remember that I took Ben’s watch to the repair store”) and recall them later by asking open-ended questions (“Alexa, where is Ben’s watch?”).