Requirements

The Watson pattern AI solution requires a range of AI components to effectively process and analyze various types of data inputs. These requirements are designed to meet specific business needs, including Retrieval-Augmented Generation (RAG), Chat (voice and text interaction), image detection, and video detection.

1. Retrieval Augmentation Generation (RAG): The AI system must be able to do text retrieval and text generation.
2. Voice and text chat: The AI system should support voice and text interaction through various channels, such as:
   1. Speech to Text conversion: Converting spoken words into written text for processing and analysis.
   2. Text to Speech conversation: The AI system must be able to process and analyze large volumes of textual data from various sources including image detection.
3. Image Detection: The AI system should include computer vision capabilities for analyzing images, such as:
   1. Object detection: Recognizing specific objects within an image.
   2. Image classification: Categorizing images based on their content or context.
4. Video Detection: The AI system must be able to process and analyze nonlive video data that's uploaded into from various sources, including:
   1. Object tracking: Following the movement of specific objects over time in a video sequence.
   2. Video analysis: Detecting patterns, trends, or anomalies within a video.

IBM watsonx services on IBM Cloud

Watsonx is IBM's next-generation AI and data platform that is designed to help businesses build, train, deploy, and scale AI models and applications, particularly in enterprise settings. Available on IBM Cloud, watsonx offers a suite of integrated tools and services that streamline the AI lifecycle—from data preparation to model training, deployment, and ongoing optimization.

Key Components of watsonx on IBM Cloud:

1. watsonx.ai: This is the core AI engine within watsonx, focused on enabling AI model development and deployment. It provides pre-built, foundation models as well as tools for training custom AI models.
2. watsonx.data: This component is IBM's AI-optimized data store designed for scalability and governed access to both structured and unstructured data.
3. watsonx.governance: This tool helps ensure that AI systems are transparent, accountable, and trustworthy. It helps with model auditing, bias detection, explainability, and compliance tracking, making it essential for regulated industries.
4. IBM watsonx Assistant for Voice: Conversational capabilities for speech and text.

For more information, see Getting started with Watson and IBM Cloud.

Maximo Visual Inspection

The IBM Maximo Visual Inspection platform that's built on cognitive infrastructure is a new generation of video and image analysis platforms. The platform offers built-in deep learning models that learn to analyze images and video streams for classification, object detection, and anomaly detection.

Maximo Visual Inspection includes tools and interfaces for anyone with limited skills in deep learning technologies. You can use IBM. Maximo Visual Inspection can be customized and deployed that demands image classification, object detection, and anomaly detection.

Maximo Visual Inspection Edge is a web-based application that you can integrate with Maximo Visual Inspection to perform AI-based inspections at the edge. Maximo Visual Inspection Edge uses data sets and trained models that are stored in Maximo Visual Inspection. In Maximo Visual Inspection Edge, you create inspections that process images from input sources.

For more information, see the following links:

1. Maximo Visual Inspection.
2. Maximo suite overview.
3. Deployment catalog tile.

This reference pattern does not describe training and fine-tuning of models, that's out of scope for this reference pattern.

IBM watsonx Assistant for Voice

IBM watsonx Assistant for Voice offers:

1. Advanced artificial intelligence technology that blends large speech models (LSMs) voice recognition, Speech to Text, and NLU capabilities.
2. Expressive voices designed to respond to customer requests in natural human-like speech with ability to understand expressions and analyze conversation sentiment. For more information, see watsonx Assistant for Voice and watsonx Assistant.

Speech to Text service

IBM Cloud Speech to Text service converts the human voice into the written word. The service uses deep-learning artificial intelligence to apply knowledge of grammar, language structure, and the composition of audio and voice signals to accurately transcribe human speech. It can be used in applications such as voice-automated chatbots, analytic tools for customer-service call centers, and multi-media transcription, among many others.

The service is available in multiple languages and is exposed as an 'http' interface and a WebSocket interface. It can be accessed by using a public or a private endpoint. For more information, see Getting started with Watson Speech to Text.