What is STVS?
STVS (Stories to Voices Studio) is an AI-enhanced computer-assisted language learning platform that integrates third-party LLM services via APIs.
It provides personalised guidance across a wide range of languages within an input–output learning cycle.
Based on major API providers’ specifications, STVS offers potential coverage of more than 140 languages.
So far, we have empirically validated STVS’s usability for 19 languages.
Why did we build STVS?
Most tools are built and tested mainly for a small set of widely taught languages. If you learn a wide variety of languages, including less commonly taught ones, you quickly find that many tools simply do not support your language well, or do not support it at all.
Many tools support the input side of learning, for example by helping you understand a text through translation, notes, or explanations, but they often stop there. Others focus on the output side, such as speaking practice. But in many output-focused tools, learners can only practise with content chosen by the platform, so the input stage offers little choice or control. In addition, in these tools, input often becomes “feedback after output”. This kind of feedback can help to some extent, but it often falls short in practice. It may be generic, it may replace your entire response with a rewritten version, or it may not clearly reflect what you were trying to say. As a result, it is hard to see what to change, where to change it, and why it matters.
STVS is built to support a wide variety of languages and bring the full input–output cycle into one place. Import a text you care about, understand it with automatic highlights, annotations, and mind maps, speak from the same content, and receive clear, traceable feedback annotated on your original wording. Questions at any step? Ask the AI tutor. Whether you’re working through intensive reading or practising speaking, you can get help instantly and keep moving forward.
How to use STVS?
Copy & Paste
Copy an article, report, document, or your own notes, then paste it into STVS.
This is the most flexible way to import content.
One-Click Import with the STVS Chrome Extension
Explore & Download the STVS Chrome Extension
The fastest, easiest way to import content—no copy and paste required.
During Intensive Reading, STVS automatically highlights key sentences and vocabulary, provides sentence-level annotations, and responds to your questions as you read. It also generates a mind map that organizes the text’s structure and main points. This mind map is saved and will be used later in the speaking module to guide your retelling and practice.
In the Speaking Practice module, STVS guides you from overall retelling to detailed description. First, STVS uses the mind map to help you deliver a logically organised overall retelling of the text. Then, STVS presents AI-generated images of key scenes or characters and asks you to describe them in detail. Across both retelling and picture description, STVS also encourages you to share your own opinions.
Your speech is automatically recorded and transcribed, and STVS annotates the transcript with personalised feedback, making areas for improvement easy to track.
STVS generates a summary to help you consolidate learning across both intensive reading and speaking practice, and monitor your progress over time.