This looks pretty exciting to me:

• the Common Voice dataset is available:https://voice.mozilla.org/data
• there’s a pre-trained model using Tensorflow: https://github.com/mozilla/DeepSpeech
• there are even bindings in Rust already: https://github.com/RustAudio/deepspeech-rs

This is super interesting. I wonder if any of this will filter into some of the older OSS projects like Harvest?