speech-to-text-wavenet

Open source speech recognition on Github: 1. speech-to-text-wavenet (https://github.com/buriburisuri/speech-to-text-wavenet) 2. kaldi(https://github.com/kaldi-asr/kaldi) 3. Speech recognition module for Python (https://github.com/Uberi/speech_recognition) 4. DeepSpeech(https://github.com/mozilla/DeepSpeech) 5. Natural Language Processing Tasks and References(https://github.com/Kyubyong/nlp_tasks) FYI.

Alternatives for the reference: https://github.com/mozilla/DeepSpeech https://github.com/pannous/tensorflow-speech-recognition https://github.com/buriburisuri/speech-to-text-wavenet

If you want to see the most basic form of this pipeline, I have a blog post on "bad speech synthesis" [0]. There are open source versions of WaveNet for TTS [1], but I have not run the code myself or seen the quality of the output. Our code for char2wav is theoretically available [2], but not yet ready for "how-to-guide" level use. [0] http://kastnerkyle.github.io/posts/bad-speech-synthesis-made... [1] https://github.com/buriburisuri/speech-to-text-wavenet [2] https://github.com/sotelo/parrot

Apparently, they tried to use the raw audio waveform with the original setup from the WaveNet paper but couldn't get it to train on their TitanX, so they used MFCCs instead. It's not exactly clear why this is the case. "Second, the Paper added a mean-pooling layer after the dilated convolution layer for down-sampling. We extracted MFCC from wav files and removed the final mean-pooling layer because the original setting was impossible to run on our TitanX GPU." [1] [1] https://github.com/buriburisuri/speech-to-text-wavenet#speec...