Pronunciation & Read-Aloud
Pronunciation & Read-Aloud
Hear the words you saved read aloud, then practice saying them with speech recognition that runs on your PC.
Standard feature — included in all 8 Practice modes.
How It Works
- A word or sentence from your Word Book is displayed
- Click the microphone button and speak
- Playto uses Whisper (via whisper-rs) to transcribe your speech locally
- Your pronunciation is scored by comparing against the expected text
Everything runs on your CPU — no GPU needed, no internet, no data leaves your PC.
Hearing words read aloud
Press the speaker icon next to a word or example sentence in your Word Book to hear it, and the Dictation practice mode plays words for you to type. Voices run on your PC and download once on first use; Japanese and Chinese have their own higher-quality voices you can download from Settings › Learning › Read aloud.
If a language has no voice yet, the speaker icon says so. See Language Support for which languages can be read aloud.
Scoring
Playto compares what it heard with the expected text and gives an accuracy score. For languages written with spaces it compares word by word, ignoring case and punctuation, and reads numbers as words (“1” matches “one”). For Japanese, Chinese and Korean it compares character by character.
Feedback: Word-by-word mismatch details showing which words were wrong and what was recognised instead.
Whisper Model
The Whisper model downloads automatically on first use. Playto tries models in order of quality:
| Model | Size | Quality |
|---|---|---|
| ggml-small.bin | ~460 MB | Best |
| ggml-base.bin | ~140 MB | Good (default) |
| ggml-tiny.bin | ~75 MB | Fast |