Overview

Amazon Lex gives application teams tools for building voice and text conversations, from speech recognition and language understanding to bot testing and deployment. It suits builders already working with AWS who need conversational interfaces across several channels. Its strongest case is the combination of AWS integrations and pay-as-you-go requests; teams should weigh language-specific feature coverage and usage charges before committing.

Key features

Design and test

Automated Chatbot Designer analyzes conversation transcripts and proposes intents and slot types, giving teams with existing conversations a starting structure instead of a blank canvas. Test Workbench adds a way to run test sets and inspect aggregate results and detailed insights. These tools support iterative bot development, though they do not remove the need to check whether the planned features work in the target language or locale.

Voice, text, and knowledge answers

Lex combines automatic speech recognition with natural language understanding for voice and text interactions. It can also answer conversational FAQs through an intent that queries an authorized knowledge source and foundation models from Amazon Bedrock. That makes it relevant for both spoken workflows and knowledge-backed questions, particularly when Bedrock is already part of the stack.

Channels and integrations

Bots can be deployed to mobile devices and chat services including Facebook Messenger, Slack, Kik, and Twilio SMS. AWS Lambda supports backend business logic, Amazon Cognito user authentication, and Amazon Polly text-to-speech. Lex also integrates natively with Amazon Connect, with pre-built integrations for other contact center solutions. This breadth is useful for AWS-centered applications, but teams should map their required channel and contact-center connections before choosing it.

Security and language coverage

Amazon Lex V2 is HIPAA eligible and AWS describes it as PCI, SOC, and ISO compliant. Clients must support TLS 1.0 or later, while AWS recommends TLS 1.2 or later. Lex V2 supports multiple languages and locales, but some features are limited to particular ones; that makes feature-by-feature language validation important for multilingual deployments.

Pricing

Lex uses pay-as-you-go request pricing, with no upfront commitment or minimum fee for the per-request examples. Text requests and speech requests are each priced at 0.00 USD per month in the plan information; the examples also give a 0.00 USD per once price for a text request and a speech request. Each text input counts as a text request, and each speech input as a speech request, so usage is tied to interaction volume rather than seats.

Automated Chatbot Designer is charged at 0.50 USD per month in the plan information, with the charge defined as $0.50 per minute of training time for transcript analysis and intent identification. Teams using it should account for training time separately from request costs.

The AWS Free Tier is 0.00 USD per free and lasts six months after account creation. New AWS customers receive up to $200 in AWS Free Tier credits, applicable to eligible AWS services including Lex; those credits must be used within 12 months of account creation. This provides a low-commitment way for eligible new customers to start, but the time window and credit expiry matter for longer evaluations.

Platforms

Lex supports Android, iOS, web, and API platforms, with multichannel deployment. The range fits teams building conversational features into applications as well as those routing bots through supported messaging services.

Who it's for

Lex is a strong fit for AWS-oriented teams building self-service voice assistants, informational or transactional bots, enterprise productivity tools, and device-control bots. It is less compelling when a team cannot rely on its AWS integrations or when required capabilities do not cover its target languages and locales.

Pros and cons

Pros: AWS Lambda, Cognito, Polly, and Amazon Connect integrations support application logic, authentication, speech output, and contact-center workflows within the AWS ecosystem.
Pros: Transcript-based bot design and test-set evaluation support both initial bot structure and ongoing assessment.
Pros: Text and speech requests are pay-as-you-go, without an upfront commitment or minimum fee in the pricing examples.
Cons: Some features are limited by language or locale, which can constrain multilingual plans.
Cons: Automated Chatbot Designer training time is charged separately, so its cost depends on training duration.

Alternatives

Consider IBM watsonx Assistant if a freemium option and self-hosted platform support suit your deployment needs; its Lite plan includes five skills per instance, with unused skills potentially deleted after 30 days of inactivity.

Chat Data offers a free plan with 100 monthly message credits, one AI agent, and 100 scraped pages, making it an option for teams whose needs fit those caps.

Botpress has a free plan capped at 25 conversations and three seats, so it may suit a small team evaluating within those limits.

Gupshup charges a standard 0.001 USD per message, with WhatsApp fees billed at actuals; consider it when that message-fee structure fits your channel requirements.

DocsBot AI has a free plan with one DocsBot, 50 source pages, 100 monthly AI credits, and one team user.

AiSensy offers a free-forever plan with fair-use limits, including up to 200 agents and 500 tags.

Workato Agentic provides 50,000 one-time credits on its free plan, a fit to compare when a one-time credit allowance meets your expected usage.

Quickchat AI has a free plan with 50 AI credits per month, one user license, and personal, non-commercial use.

For more options, browse Chatbot builders and Conversational AI Platforms.

Verdict

Choose Amazon Lex if your team wants voice and text bots integrated with AWS services and needs multichannel deployment with usage-based request pricing. Its strongest reason to choose is the AWS-connected path from bot design through application logic and contact-center integration; look elsewhere if language-specific feature coverage or the separate training-time charge does not fit your plan.

	Amazon Lex	CustomGPT.ai
Free plan	Yes	No
Free trial	No	No
Paid from	$0.50/mo	$99/mo
Open source	No	No
Platforms	Android, api, iOS, Web	Web
Free plan	No	No
Deployment channels	multichannel	multichannel
Visual flow builder	Yes	—
AI bot generation	Yes	Yes
Knowledge-base answers	Yes	Yes
Human handoff	Yes	Yes
Paid from	—	$89/mo

Amazon Lex vs CustomGPT.ai

Amazon Lex

CustomGPT.ai