We've written previously about how machine learning – also sometimes called artificial intelligence – is moving out of the research lab and into the workplace to help solve real business needs. There's no denying, however, that implementing machine learning still involves quite a bit of expertise and skill. That's why Google has created a number of APIs to let you access “pre-baked” machine learning systems that are already trained to handle a range of tasks.
Google Cloud Machine Learning APIs
Using Google Cloud Machine Learning APIs, you can immediately start applying machine learning to:
- Images: using the Google Cloud Vision API, you can automatically tag images based on their content – which can include individual objects faces or any printed words – and then let your users search for images with specific content.
- Video: using the Google Cloud Video Intelligence API, you can automatically tag every frame of a video archive with keywords and annotations, letting users find the specific segments of the video files in your catalogue that meet their needs.
- Speech: using the Google Cloud Speech API, you can accurately and automatically convert audio to text in more than 80 languages and variants, and use the results to handle voice control of applications as well as provide transcription.
- Language: using the Google Cloud Translation API, you can get fast dynamic translation of text between any two supported languages – quickly enough to translate websites and applications “on the fly”.
- Text: using the Google Natural Language API, you can automatically tag and analyse the content of news articles, blog posts, conversations on social media sites or in messaging apps, or any other natural language text. You can use the results to understand what people feel and are saying about your organisation; or to route and prioritise customer contacts; or simply to help users find relevant articles.
- Jobs: using the Google Cloud Jobs API, you can provide users with a highly intuitive approach to finding a new job that anticipates what they're looking for and recommends targeted opportunities.
If your needs aren't covered by any of these APIs, you can take advantage of open source models developed by members of Kaggle, the world's largest community of data scientists, now part of the Google family, or build your own model using Google's Large Scale Machine Learning Service. This provides all the tools you need to build your own sophisticated, large scale machine learning models.
If you’d like to find out more about how machine learning could help your organisation, please call us to talk to one of our Google Cloud Platform specialists.