aip_trainer/lambdas/app_description.md

AI Pronunciation Trainer

See my fork of AI Pronunciation Trainer repository for more details.

Selecting a new example invalidates all previous inputs (Text-to-Speech audio and Speech-to-Text recordings): in that case you should record again your audio speech.

Right now this tool uses {sample_rate_start} as sample rate value. From empirical tests the best sample rate value is 48000.