Learn core Google AI and machine learning tools
Looking to infuse AI and machine learning into your cloud apps? Use this list of terms to explore which Google cloud services offer features for speech to text, image analysis and more.
Artificial intelligence isn't science fiction anymore. For some enterprises, the technology already provides many benefits. With machine learning algorithms, for example, applications can "learn" from and predict possible outcomes based on ever-growing data sets. Top cloud vendors, including Google, now offer various AI and machine learning tools for the enterprise.
The Google Cloud machine learning and AI suite includes AutoML services to train models, as well as various APIs that can analyze and translate text, perform image and speech recognition, and search videos with metadata. Enterprises now have access to technology that Google uses for its own applications, such as the image search capabilities used for Google Photos.
There's also Google Cloud AI Platform, a code-based development environment that enables enterprises to create and deploy machine learning applications. It supports Kubeflow, a machine learning toolkit for Kubernetes.
Some other key Google AI and machine learning tools follow.
Google Cloud Natural Language
Google Cloud AutoML Natural Language and the API both work to understand sentiment and customer conversations through text. The API uses five methods to analyze and annotate text: sentiment, entity, entity sentiment, content classification and syntax analysis. These methods inspect text for tone, proper and common nouns, attitude towards an entity and linguistic information.
Google Cloud Speech
Google Cloud has two speech APIs: Speech-to-Text and Text-to-Speech. These services convert audio to text, and vice versa, in real time. Both use neural network algorithms and models to complete conversions. Speech-to-Text recognizes over 120 languages and various dialects. It also has voice command-and-control capabilities. Text-to-Speech can produce over 100 synthesized human voices and supports over 20 languages and variants.
Google Cloud Dialogflow Enterprise Edition
Dialogflow is a platform that developers can use to create and deploy a conversational UI for various types of apps and devices, such as a chatbot or interactive voice response system. It can use various APIs, such as Google Cloud Speech-to-Text and Text-to-Speech, and can engage with users in more than 20 languages.
Google Cloud Translation
AutoML Translation and the API can identify and translate text for numerous languages. The API uses Google's neural machine translation capabilities to quickly translate text. The AutoML option automatically trains custom translation models.
Google Cloud Vision AI
Google Cloud Vision includes two AI services. AutoML Vision automates training for custom machine learning models that can classify images based on custom-defined labels. The Vision API offers pretrained models that analyze and categorize images based on predefined labels, as well as detection features for facial characteristics and landmarks.
Google Cloud Video Intelligence
This service again comes in two options: AutoML Video Intelligence and the API. The AutoML offering trains machine learning models with classification capabilities for video shots and segments. The Cloud Video Intelligence REST API comes with pretrained models to detect places and objects in video, as well as shot changes and explicit content.
Google Cloud Recommendations AI
Recommendations AI is a managed service that helps enterprises provide personalized suggestions to their customers, based on a provided product catalog and customer behavior. Developers can customize the service depending on the outcome they desire, such as increased revenue or engagement.
Google Cloud AI Hub
AI Hub is a private, hosted repository of AI resources, such as algorithms and TensorFlow modules, with sharing capabilities. It also gives enterprises access to Google's machine learning models and research papers, as well as public material from third-party providers.
Google Cloud AutoML Tables
Developers and data scientists can use Google Cloud AutoML Tables to build and deploy predictive machine learning models based on structured data. Common use cases include models that can help predict customer demand based on spending habits or foresee the likelihood of fraud in financial intuitions.
Google Cloud AI Platform Notebooks
AI Platform Notebooks is a managed notebook service with a JupyterLab interface. It integrates with BigQuery, Cloud Dataproc and Cloud Dataflow, as well as AI Hub. It is preinstalled with machine learning libraries, such as TensorFlow, PyTorch and scikit-learn.
Google Cloud AI Platform Deep Learning VM Image
Deep Learning VM Image is a series of Compute Engine VM images specifically tailored for machine learning workloads. The images are preinstalled with deep learning and machine learning frameworks -- such as TensorFlow, PyTorch and scikit-learn -- and admins can add Cloud TPU and GPU support.
Google Cloud TPU
The Tensor Processing Unit (TPU) is an application-specific integrated circuit that can handle the computational needs of machine learning applications. With Google Cloud TPUs, enterprises can more quickly train large and complex machine learning models. Google offers three versions of its TPUs, with up to 11.5 petaflops and 4 TB of high-bandwidth memory.
Are enterprises ready for machine learning and AI services?
Get familiar with Google's big data services
L'Oréal to 'revolutionize' beauty services using Google Vertex