Proven tools for proven results
Create intelligent IVR, chatbot, and messaging experiences with intuitive tools built on Nuance speech and AI technologies, APIs, and microservices.
Mix Dialog
Design, develop and test multi‑language, omni‑channel conversational AI experiences in a single project across voice and digital channels—all with an easy to use graphical UI.
Build on Conversational AI
Today’s consumers want choice and flexibility: Choice of channels, and flexibility in the way they interact. Mix.dialog lets designers and developers build on a set of predefined dialog nodes that incorporate conversational AI best practices and cover use cases from comprehensive FAQ‑type VAs to highly personalized, transactional bots and IVRs—without having to write a single line of code.
Build for voice and digital channels
Build conversational AI experiences for voice‑enabled and digital channels in a single project. Optimize the logic for each channel and modality while ensuring consistency and re‑use within a single project. Leverage rich text, media and action buttons in digital channels, and Nuance Vocalizer TTS and prerecorded prompts in voice channels.
Leverage context and session awareness
Conversational logic built in Mix.dialog builds on context and session awareness. User messages such as “Am I due for a new phone yet?” will be interpreted along with information from prior messages in the same dialog to provide the right answer. Take the dialog understanding further by including external data such as location, account details, and customer preferences to create more personalized experiences.
Mixed‑initiative dialogs
In Mix.dialog, call flow designers build on core components that can orchestrate mixed‑iniative dialogs. First time users will be guided by the system through conversations step by step, while more experienced users can take the fast track and take control of the conversation.
Multi‑intent
Bots built with Mix know how to multitask. Users can start a conversation and go down a path for one intent; and when they add a second intent, they can branch out of the current flow to resolve it—and then resume the original task seamlessly.
Develop and integrate
Mix.dialog lets developers integrate with existing backend systems for authentication, data access, and transaction fulfillment.
Integrate with the Mix.dialog Runtime API
Mix.dialog runtime APIs combine Nuance's speech technologies, NLU, and dialog orchestration into a single API—allowing for easier integration with clients including IVR, VAs or bots, smart speakers, mobile apps, and social media engagement channels.
Back‑end Integration
Integrate external systems (such as CRM, customer profile) into your call flow logic to personalize the user experience and better fulfill their needs.
Mix.dialog allows direct integration with RESTful web services and client-side data integration which help create custom client integrations for mobile and web applications that can leverage access to client‑side data and logic.
Build dialog flow logic without coding
Define personalized call flow logic and business rules without having to write a single line of code. Build complex, nested conditions via an easy‑to‑manage, form‑filling approach. Leverage data objects with schemas mapping to your back‑end methods to minimize data manipulation requirements.
Test
Use built‑in validation and the Try mode to test your dialog logic before pushing it into production. Drive the application logic through all possible paths by simulating different user messages and back‑end response data.
Test drive the application logic
The Try mode in Mix.dialog allows developers to test drive the application logic without having to deploy to the target environment. Simulate text and voice modalities and verify system responses, while getting visual feedback on the call flow execution in a protected environment.
Simulate Backend Responses
While test‑driving application logic through all possible paths in Try mode, simulate backend responses by entering different response data right into the test window.
Built‑in Validation
When configuring call flow logic details, Mix.dialog supports developers by validating the configured dialog nodes and points them at missing content or flawed logic.
Mix NLU
Build natural language processing domains and continuously refine and evolve your NLU model based on real‑world usage data. Define user intents ('book a flight') and entities ('from JFK to LAX next Wednesday') and provide sample sentences to train the DNN‑based NLU engine.
Define Intents and Entities(Open a new window)
Train your NLU model with sample phrases to learn to distinguish between dozens or hundreds of different user intents. For each intent, define the entities required to fulfill the customer request. Create custom entities based on word lists and everyday expressions or leverage ready‑made entities for numbers, currency, and date/time that understand the variety of ways customers express that information.
Test the Model(Open a new window)
Train the NLU model at any time and test it against practice sentences. Identify problem areas where intents overlap too closely, confidence levels need to be boosted, or additional entities need to be defined.
NLU and Speech Recognition Tuning
Deploy the trained NLU model both to the NLU engine and at the same time, as a domain language model, to the speech‑to‑text transcription engine. This provides the highest accuracy in speech recognition results, semantic parsing, and understanding of user utterances based on your application’s specific language domain.
Mix Dashboard
Use one central view for managing users, access rights, and project versions and deployments. The Mix Dashboard also allows for promotion flows from a sandbox environment to staging and production environments while letting you control multi‑datacenter, multiregional and hybrid deployment models.
Organize projects and manage access (Open a new window)
Manage conversational AI projects across your enterprise. Define the supported languages, channels, and modalities on a per‑project basis.
Promotion flows (Open a new window)
Control project versions by integrating with common version management systems. Define deployment packages, and promote them from a sandbox environment to staging and production environments across multiple regions or data centers. Roll back immediately in case of problems with a new version.
Manage Project Settings(Open a new window)
Add additional languages or additional channels and modalities to your projects.