V 0.2.0 is based on deepspeech 2?

[shaguftamethwani]

Is V 0.2.0 following DeepSpeech 2 implementation? Since there’s streaming support it means bidirectional RNNs are not used?

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

Is V 0.2.0 following DeepSpeech 2 implementation? Since there’s streaming support it means bidirectional RNNs are not used?

This is not DeepSpeech 2 implementation, and yes, no more bidirectional layer.

[shaguftamethwani]

@lissyx Is there any blog post about the new architecture? Like in the first one there were nice diagrams and posts describing it…

[lissyx ((slow to reply) [NOT PROVIDING SUPPORT])]

https://hacks.mozilla.org/2018/09/speech-recognition-deepspeech/