Re:Speech Recognition Is Not Solved
by dharma1 on Tuesday October 24, 2017 @03:54PM
(#15544715)
You're right in that TPU's allow Google to train very large datasets faster and using less power. But I think a reasonable ASR model should be trainable with GPUs alone.
The issue previously has been a lack of large enough high quality annotated datasets, and open source ASR libraries being a bit behind or not well integrated with cutting edge deep learning. I think that's changing now though. I hope it won't take too long until pre-trained, reasonable size and high accuracy TensorFlow/Kaldi models for many languages are common.
The issue previously has been a lack of large enough high quality annotated datasets, and open source ASR libraries being a bit behind or not well integrated with cutting edge deep learning. I think that's changing now though. I hope it won't take too long until pre-trained, reasonable size and high accuracy TensorFlow/Kaldi models for many languages are common.