Glosso Studio
An Android-first English course that measures pronunciation phoneme by phoneme. In open beta: free to use, no ads, no third-party trackers. Cloud scoring by default, or an optional 342 MB on-device model that keeps audio local.
Most pronunciation apps push your speech through speech-to-text and tell you whether the words matched. That says almost nothing about how you sound. Glosso gives a verdict on each individual sound: which phonemes you hit, which drifted, and where inside the word the problem is.
It is Android-first, English-first, and in open beta. No ads and no payment while the beta lasts.
What it does
- Records a sentence and scores each phoneme on its own.
- Names the sound that slipped: for example, /t/ where the word needed /θ/.
- Shows the IPA of what it heard, next to a reference pronunciation.
- Builds the next drill from your own errors and tracks each sound over time.
Where scoring runs
- Cloud by default. A normal install sends the recording over HTTPS to Glosso’s own server, where it is scored in memory and discarded as soon as the request ends. It is never stored, sold or used to train models.
- Optional on-device. Settings offers a one-time download of an acoustic model (about 342 MB). With it installed, scoring runs entirely on the phone and works offline. Delete it any time to go back to cloud scoring.
There is no mandatory account and no sign-up wall: progress stays on the device. No advertising or third-party tracking SDKs are in the build.
Honest limits
If the recording has no usable speech, or is too noisy or too quiet, Glosso says the attempt could not be evaluated instead of showing 0/100. The score is an estimate, not a certificate; even the reference recordings average around 89 out of 100.
Stack
One Kotlin Multiplatform engine does the phonetics, shared between the Android app and the scoring service so both paths agree. Jetpack Compose for the app, ONNX Runtime with a wav2vec 2.0 acoustic model for inference, Ktor for the service, Room for local state. Android 8.0 or newer. English first, against a British reference accent; French is next.
Beta
Glosso is in open beta: you can use it free of charge while it keeps being built. The beta stays open until , and during this stage you can try what is available, look at your pronunciation, and help decide what works and what should improve. Your feedback is part of how Glosso develops.
Beta Founder. Everyone who joins during the beta becomes a Beta Founder, permanently, even if they later change plan. When the beta ends you keep that status and get special benefits, including special terms for Glosso Pro. The exact benefits are not fixed yet; they will be defined during the beta. Nothing extra to do: if you join during the beta, you are a Beta Founder.
After the beta. The end of the beta does not mean Glosso stops being free. From 2028 it moves to three tiers: Free (a free experience so anyone can learn), Pro (more advanced features and analysis) and Personalised (practice adapted to each student from their own results). The details will be decided during the beta, based on how the product is used and on user feedback.
Glosso is a proprietary product. It grew out of an earlier offline phonetic prototype, but the current app is a separate codebase.
Glosso Beta · Free until · Beta Founder from day one.
Written with AI assistance. I plan to rewrite it by hand.