Analysis updated 2026-05-18
Practice all four parts of the C1 Advanced speaking exam with an AI examiner.
Get feedback with transcript citations and acoustic analysis after a session.
Track recurring mistakes over time with personalized follow-up exercises.
| scastarnado/englishspeakerteacherorsomethinglikethat | 00kaku/gallery-slider-block | 3rd-eden/ircb.io | |
|---|---|---|---|
| Stars | 0 | — | — |
| Language | JavaScript | JavaScript | JavaScript |
| Last pushed | — | 2021-05-19 | 2016-11-16 |
| Maintenance | — | Dormant | Dormant |
| Setup difficulty | moderate | easy | easy |
| Complexity | 3/5 | 2/5 | 2/5 |
| Audience | general | general | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires installing Ollama and downloading a multi-gigabyte local AI model before first use, Windows only.
C1 Speaking Coach is a desktop application for practicing the C1 Advanced English speaking exam, built to run entirely on the user's own computer using local AI models rather than a cloud service. The README is explicit that the app is an independent practice tool and is not affiliated with or endorsed by Cambridge, the organization behind the real exam. The app simulates all four parts of the C1 speaking exam with realistic timing, using an AI interlocutor that acts like an examiner and an AI co-candidate for the discussion sections. After a practice session it gives feedback that cites the actual transcript along with acoustic analysis, and it tracks a learner's progress over time to spot recurring mistakes and suggest personalized exercises. Because everything runs locally, no recordings or data are ever uploaded, and once set up the app works fully offline. To run the local AI, the app depends on a separate tool called Ollama, which downloads and runs open language models on the user's own machine. The README walks through two installation paths in a lot of beginner-friendly detail: installing a ready-made Windows app for people who just want to use it, or installing from source with Git, Node.js, and Python for people who want to modify the code or build their own installer. Model sizes range from a smaller 3 billion parameter model for lower-spec machines up to a 14 billion parameter model for computers with more RAM and a strong GPU. System requirements scale with the chosen model, starting at a four-core CPU with 8 gigabytes of RAM and no GPU for the low-resource profile, up to 32 or more gigabytes of RAM and a GPU with 12 or more gigabytes of video memory for the high-quality profile. Currently only Windows is supported, with Linux and macOS listed as planned. The installed app checks GitHub periodically for new releases and can update itself in place while preserving saved data.
C1 Speaking Coach is a local-first desktop app that simulates the C1 Advanced English speaking exam using AI examiners and gives progress feedback.
Mainly JavaScript. The stack also includes JavaScript, Node.js, Python.
The README does not state a license for this code.
Setup difficulty is rated moderate, with roughly 1h+ to a first successful run.
Mainly general.
This repo across BitVibe Labs
Verify against the repo before relying on details.