Amazon Lex FAQs

Amazon Lex is a service for building conversational interfaces using voice and text. Powered by the same conversational engine as Alexa, Amazon Lex provides high quality speech recognition and language understanding capabilities, enabling addition of sophisticated, natural language ‘chatbots’ to new and existing applications. Amazon Lex reduces multi-platform development effort, allowing you to easily publish your speech or text chatbots to mobile devices and multiple chat services, like Facebook Messenger, Slack, Kik, or Twilio SMS. Native interoperability with AWS Lambda and Amazon CloudWatch and easy integration with many other services on the AWS platform including Amazon Cognito, and Amazon DynamoDB makes bot development effortless.

To start using Amazon Lex, simply sign into the AWS Management Console and navigate to “Lex” under the “Artificial Intelligence” category. You must have an Amazon Web Services account to start using Amazon Lex. If you do not already have one, you will be prompted to create one during the sign-up process. Please refer to the Amazon Lex V2 Getting Started Guide for more information.

The most common use-cases include:

• Self-service voice assistants and chatbots – build a call center bot

• Informational bot – build an automated customer support agent or bot that answers questions

• Application/Transactional bot – build a stand-alone pizza ordering agent or a travel bot

• Enterprise Productivity bot – build custom bots to connect to enterprise data resources

• Device Control bot– use Amazon Lex to issue control commands to connected devices

Amazon Lex leverages AWS Lambda for Intent fulfillment, Amazon Cognito for user authentication, and Amazon Polly for text to speech.

No machine learning expertise is necessary to use Amazon Lex. Developers can declaratively specify the conversation flow and Amazon Lex will take care of the speech recognition and natural language understanding functionality. Developers provide some sample utterances in plain English and the different parameters (slots) that they would like to collect from their user with the corresponding prompts. The language model gets built automatically.

For a list of the supported Amazon Lex AWS regions, please visit the AWS Region Table for all AWS global infrastructure. Also for more information, see Regions and Endpoints in the AWS General Reference.

Amazon Lex scales to your needs and does not impose bandwidth constraints.

Amazon Lex is a completely managed service so you don’t have to manage scaling of resources or maintenance of code. Your interaction schema and language models are automatically backed up. We also provide comprehensive versioning capability for easy rollback. Amazon Lex architecture does not require storage or backups of end user data.

Amazon Polly converts text inputs to speech. Amazon Lex is a service for building conversational interfaces using voice and text.

Yes. Amazon Lex uses deep learning to improve over time.

To create a bot, you will first define the actions performed by the bot. These actions are the intents that need to be fulfilled by the bot. For each intent, you will add sample utterances and slots. Utterances are phrases that invoke the intent. Slots are input data required to fulfill the intent. Lastly, you will provide the business logic necessary to execute the action. An Amazon Lex bot can be created both via Console and REST APIs.

Yes. Amazon Lex provides the option of returning parsed intent and slots back to the client for business logic implementation.

Amazon Lex provides deep integration with AWS Lambda and you can validate user input using the initialization and validation codeHook. This code gets executed at every turn of the conversation. The codehook can be used to set up session parameters, validate user input and customize responses.

To build an Amazon Lex bot, you will need to identify a set of actions - known as 'intents’ -- that you want your bot to fulfill. A bot can have multiple intents. For example, a ‘BookTickets’ bot can have intents to make reservations, cancel reservations and review reservations.

An ‘utterance’ is the spoken or typed phrase to invoke an intent. For example, to invoke the intent to make reservations, you would provide a sample utterance such as, “Can I make a reservation?”

To fulfill an intent, the Amazon Lex bot needs information from the user. This information is captured in ‘slots’. For example, you would define show name and time as slots for intent to make reservations.

Amazon Lex elicits the defined ‘slots’ by using the ‘prompts’ provided. For example, to elicit value for the slot ‘time’ you will define a prompt such as “What show time would you like to reserve?”. Amazon Lex is capable of eliciting multiple slot values via a multi-turn conversation.

Amazon Lex integrates with AWS Lambda for ‘fulfillment’ of the action or business logic. Alternately, you can configure Amazon Lex to return parsed intent and slot values to the client for action fulfillment.

You can track metrics for your bot on the ‘Monitoring’ dashboard in the Amazon Lex Console. Currently, you can track the number of missed utterances, request latency and traffic by channel for your bot. You can view list of utterances that were not recognized by your bot, aka 'missed utterances'. With these monitoring capabilities, you view how your users are interacting with the bot and make improvements over time.

The automated chatbot designer helps you create a bot design in just a few clicks. You first provide a link to the S3 location that contains your conversation transcripts via the Lex Console (or the SDK). The automated chatbot designer will then process these transcripts to surface a chatbot design that includes user intents, sample phrases associated with those intents, and a list of all the information required to fulfill them. You can then review the results provided by the automated chatbot designer and add the intents and slot types that are best suited to your bot.

The transcripts must contain conversations between a caller and an agent in standardized JSON format. You can find a sample transcript in this format in Amazon Lex documentation. Amazon Connect customers using Contact Lens can directly use conversation transcripts in their original format. Conversation transcripts from other transcription services may require a simple conversion. You can find details about the conversion process here.

All English locales (US, UK, AU, IN, SA) supported by Amazon Lex are supported by the automated chatbot designer. At preview, the automated chatbot designer supports US English.

Building a bot triggers machine learning and creates the models for your bot. A new version of your intents and slot types is created. Once created a version is immutable.

You can test your Amazon Lex bot via the test window on the console. Any business logic implemented in AWS Lambda can be tested via this console as well. All supported browsers allow for testing text with your Amazon Lex bot; voice can be tested from a Chrome browser.

Amazon Lex provides SDKs for iOS and Android. You can develop bots for your mobile use cases with these SDKs. User authentication can be enabled via Amazon Cognito.

Amazon Lex bots can be published to messaging platforms like Facebook Messenger, Slack, Kik, and Twilio SMS. To publish the bot you can provide the tokens for authentication in the console, and we will store it securely and provide a callback URL that you can provide to the chat service.

You don’t need to certify your bot with Amazon prior to deployment.

Yes. You can build and deploy a version of your bot into production while you continue to develop on a different version. Every version of an Amazon Lex bot will have an ARN. Each version can be associated with a different alias. You can use these tools to set up dev, stage and prod environments.

Yes. You can deploy a specific version to each messaging service. Every version of Amazon Lex will have an ARN. Each version can be associated with an alias. You can use different aliases for deployment to different messaging service. Also, you can have multiple bots deployed to the same messaging service.

Amazon Lex bots are designed for a request and response interaction or a continuous streaming conversation. With the request and response interaction, Amazon Lex supports up to 15 seconds of speech input. In a streaming conversation, all user inputs across multiple turns are processed in one streaming API call up to 15 seconds of speech input including silence.

Yes, you can just choose the PostContent API to provide voice input and choose text output.

Currently, Amazon Lex supports US English, Spanish, French, German, Italian, Japanese, Australian English, British English, Canadian French, Latin American Spanish, and US Spanish.

Amazon Lex supports the following formats for input audio: LPCM and Opus; Supported output audio formats: MPEG, OGG, PCM.

Amazon Lex can be accessed from VPC via public endpoints for building and running a bot. Currently, Amazon Lex does not provide a VPC endpoint.

No. End users will need to access the Amazon Lex runtime endpoint over the Internet.

Lex V2 console and APIs use an updated information architecture (IA) to deliver simplified versioning, support for multiple languages in a bot, and streaming capabilities. Additional improvements include the saving of partially completed bot configurations, renaming of resources, simplified navigation, bulk upload of utterances, and granular debugging.

You can use the streaming API to conduct a continually streaming conversation with a Lex bot. With streaming conversation, the bot continuously listens and can be designed to respond proactively to user interruptions and pauses. For example, you can configure the bot to keep a conversation going when a user needs more time to respond by sending periodic messages such as “Take your time. Let me know once you are ready.”

Amazon Lex bots are designed for a request and response interaction or a continuous streaming conversation. With the request and response interaction, each user input (voice or text) is processed as a separate API call. In a streaming conversation, all user inputs across multiple turns are processed in one streaming API call. Please refer to the Amazon Lex pricing page for more details.

Yes, Amazon Connect contact flows work with both Lex V2 and V1 APIs. You can use the Lex V2 console to create and integrate bots with Amazon Connect.

No. If you want to take advantage of V2 features, you will need to recreate your bot with V2 APIs. The Lex V1 APIs are not compatible because V2 APIs use an updated information architecture to enable simplified resource versioning and support for multiple languages within a bot. Converting to V2 APIs is easy, so get started with this step by step migration guide.

The Amazon Lex V2 APIs and enhanced console experience is available in all existing 8 regions and languages including US English, Spanish, French, German, Italian, Japanese, Australian English, British English, Canadian French, Latin American Spanish, and US Spanish. For a list of the supported Amazon Lex AWS regions, please visit the AWS Region Table.

No. These features are only available in the V2 APIs. If you want to take advantage of these features, you can migrate to V2 APIs by following this migration guide.

Yes, you can access the V1 console in the AWS Management Console. Once in the Lex console, you can navigate between the V1 and V2 console. The bots created in the V1 console will only be visible within the V1 Console. You will not be able to access your V1 bots in the V2 console until you recreate them in the V2 console. Migrating your bots to V2 is easy, here is a step by step migration guide.

You can click on the link in the left navigation bar to choose V1 or V2 as your console.

Yes. The existing Lex V1 APIs are still supported. You can continue to use them to build and conduct your bot conversations.

Alexa Skills Kit (ASK) is used to build skills for use in the Alexa ecosystem and devices and lets developers take advantage of all Alexa capabilities such as the Smart Home and Flash Briefing API, streaming audio and rich GUI experiences. Amazon Lex bots support both voice and text and can be deployed across mobile and messaging platforms.

Amazon Lex does not support wake word functionality. The app that integrates with Amazon Lex will be responsible for triggering the microphone, i.e. push to talk.

Currently we do not support the Alexa voice for Amazon Lex responses. However, there are 7 other voices from which to choose.

Amazon Lex provides the ability for you to export your Amazon Lex bot schema into a JSON file that is compatible with Amazon Alexa. Once downloaded as JSON, you need to log in to the Alexa developer portal, navigate to the ‘Interaction Model’ tab, launch the Alexa Skill Builder, and paste the bot schema into the Code Editor of your Alexa Skill.  More details and steps can be found in the Amazon Lex documentation.

No. Only the bot definition will be downloaded.

All Alexa platforms that support Alexa skills can be used: The Amazon Echo, Amazon Dot, Amazon Look, Amazon Tap, Amazon Echo Show and any third-party Alexa-enabled devices.

Amazon Lex may store and use voice and text inputs processed by the service solely to provide and maintain the service and to improve and develop the quality of Amazon Lex and other Amazon machine-learning/artificial-intelligence technologies. Use of your content is necessary for continuous improvement of your Amazon Lex customer experience, including the development and training of related technologies. We do not use any personally identifiable information that may be contained in your content to target products, services or marketing to you or your end users. Your trust, privacy, and the security of your content are our highest priority and we implement appropriate and sophisticated technical and physical controls, including encryption at rest and in transit, designed to prevent unauthorized access to, or disclosure of, your content and ensure that our use complies with our commitments to you. Please see https://aws.amazon.com/compliance/data-privacy-faq/ for more information. You may opt out of having your content used to improve and develop the quality of Amazon Lex and other Amazon machine-learning/artificial-intelligence technologies by using an AWS Organizations opt-out policy. For information about how to opt out, see Managing AI services opt-out policy.

Yes. You can request deletion of voice and text inputs associated with your account by contacting opting out. Deleting voice and text inputs may degrade your Amazon Lex experience. For information about how to opt out, see Managing AI services opt-out policy.

Only authorized employees will have access to your content that is processed by Amazon Lex. Your trust, privacy, and the security of your content are our highest priority and we implement appropriate and sophisticated technical and physical controls, including encryption at rest and in transit, designed to prevent unauthorized access to, or disclosure of, your content and ensure that our use complies with our commitments to you. Please see https://aws.amazon.com/compliance/data-privacy-faq/ for more information.

You always retain ownership of your content and we will only use your content with your consent.

Any content processed by Amazon Lex is encrypted and stored at rest in the AWS region where you are using Amazon Lex. Some portion of content processed by Amazon Lex may be stored in another AWS region solely in connection with the continuous improvement and development of your Amazon Lex customer experience and other Amazon machine-learning/artificial-intelligence technologies. You can request deletion of voice and text inputs associated with your account by contacting AWS Support. Your trust, privacy, and the security of your content are our highest priority and we implement appropriate and sophisticated technical and physical controls, including encryption at rest and in transit, designed to prevent unauthorized access to, or disclosure of, your content and ensure that our use complies with our commitments to you. Please see https://aws.amazon.com/compliance/data-privacy-faq/ for more information.

Yes, subject to your compliance with the Amazon Lex Service Terms, including your obligation to provide any required notices and obtain any required verifiable parental consent under COPPA, you may use Amazon Lex in connection with websites, programs, or other applications that are directed or targeted, in whole or in part, to children under age 13. Amazon Lex does not store or retain voice or text utterance information from websites, programs, or applications that are identified by customers in accordance with the Amazon Lex Service Terms as being directed or targeted, in whole or in part, to children under age 13 and subject to COPPA.

For information about the requirements of COPPA and guidance for determining whether your website, program, or other application is subject to COPPA, please refer directly to the resources provided and maintained by the United States Federal Trade Commission. This site also contains information regarding how to determine whether a service is directed or targeted, in whole or in part, to children under age 13. whole or in part, to children under age 13.

Amazon Lex currently supports SDKs for runtime services. IoS and Android SDKs, as well as Java, JS, Python, CLI, .Net, Ruby, PHP, Go, and CPP support both text and speech input.

You can build bots using SDKs: Java, JavaScript, Python, CLI, .NET, Ruby on Rails, PHP, Go, and CPP.

Depending on your AWS support contract, Amazon Lex is supported under Developer Support, Business Support and Enterprise Support plans.  You can also post your queries on the Amazon Lex forums.

Every input to an Amazon Lex bot is counted as a request. For example, if an end user provides 5 inputs to the bot as part of conversation, these are billed as 5 requests. Usage is metered and billed per request.

Get started with the AWS Free Tier. Please see the Amazon Lex Pricing Page for current pricing information.

Yes. You can try Amazon Lex for free. From the date you get started with Amazon Lex, you can process up to 10,000 text requests and 5,000 speech requests per month for free during the first year.