We have just released 10 new AI-based extensions in the Directus Marketplace leveraging the very best-in-class offerings. Our extensions exist in three categories – data transformation, data generation, and data analysis – each bringing new capabilities to Directus Automate.
Transcribe audio into a formatted transcript
The AI Transcription extension brings Deepgram’s powerful speech-to-text API directly to Directus Automate. Provide a file in your project and immediately receive a formatted transcript, along with timestamps at a paragraph, sentence, and word level.
Optionally transcribe asynchronously, enable speaker detection, or boost brand or specialized keywords.
Use cases include improving accessibility of media and analytics of large datasets of voice recordings.
Translate content into 30+ languages
The AI Text Translation extension enables DeepL’s translation services for over 30 languages. Provide text and a target language, and the operation will detect the source language and provide a translation.
You can combine this with the built-in translations interface to automatically provide multilingual content in your applications.
Automatically generate alt text for images
The AI Alt Text Writer extension uses Clarifai’s image recognition models to describe the content of an image.
While we believe alt text is best written by content authors, being able to generate it can still improve accessibility of your applications.
Extract text from within images
The AI Text Extraction extension uses Clarifai’s image OCR models to extract text in your images.
You will receive information about different areas that contain text, as well as an overall output. This can be used any time important data is locked in images – receipts, business cards, signs, and more.
Generate text with a prompt and input
The AI Writer extension uses OpenAI’s GPT models to create new text based on a prompt and an input. Built-in prompts include writing social posts, generating SEO descriptions, changing the length of the provided text, fixing spelling and grammar, and more.
You can also provide a custom prompt or enter advanced mode to provide multi-step prompts which make this a flexible and powerful operation.
Generate new images with DALL-E
The AI Image Generation extension uses OpenAI’s DALL-E models to generate new images based on a prompt. You can also specify quality and size to balance output and cost.
This is useful for ideating physical products, creating mockups and generating social or meta images.
Generate realistic speech from text
The AI Speech Generation extension uses LOVO’s Genny service to create realistic speech clips from provided text. You can pick any of the provided speakers to create an output that fits your needs.
This can be used for at-scale ad or video voiceover generation.
Get your own personal AI text analyzer
The AI Text Intelligence extension uses Deepgram’s models to analyze text to help you understand intents, sentiment, topics, and generate summaries.
Quickly understand the meaning behind text, and combine with the AI Transcription operation to understand the meaning behind speech.
Automatically detect focal points in images
The AI Focal Point Detection extension uses OpenAI’s vision models to determine the main point of interest in an image.
In Directus 10.9 we released focal point support, allowing for cropping around a specific point in an image. This operation allows for automatic detection of this point, ensuring your subject is always visible.
Moderate images with AI
The AI Image Moderation extension uses Clarifai’s classifier models to analyze images for drugs and suggestive or explicit material, and gore.
You can use this operation to automatically flag or block images that aren’t wanted by your team.
Explore and Extend
Directus is the toolkit for building applications. It’s modular. It’s powerful. And it places you in control. And this is about providing tools to use, if you want, when you want, that weren’t there before.
We hope you find value in these new extensions in your projects.