About Sada
Sada is a free website that helps you speak English like real speakers: pick a video, say its lines with the speaker or after them, and see which of your words were heard and how far behind you were. It exists because “repeat after me” practice is everywhere, but people practising alone can’t tell what they got wrong.
Why Sada exists
Many learners shadow YouTube videos, listening and repeating without knowing which words came out clearly and which were lost. Sada adds what that practice is missing: a score after every take, your recording next to the original, and a history that shows your first try against your latest. It is completely free, with no account and no ads.
How the score is worked out
- A speech-recognition model running on your device listens to your take and writes down the words it heard.
- Those words are compared with the video’s captions word by word, showing what was heard, missed or added.
- Words that are hard to hear in the video itself are not held against you.
- Timing comes from the word timings in automatic captions, so you see how far behind the speaker you were in Shadow mode, or your pace in Repeat mode.
What Sada does not measure
Sada does not grade each individual sound, and it does not judge your accent. The score says whether your words came across as the speaker’s did; judging the accent is for your ear, listening to your take next to the original. A word can also be missed because of noise or the microphone rather than your pronunciation.
Your data
Your recordings, scores and library are kept in your browser and never uploaded. The server only sees which video you ask captions for, and keeps anonymous daily counts of how many people start practising. Details are on the privacy page.
Sada is an independent project, not affiliated with YouTube, Google, or the channels and speakers whose videos you practise with.