From 05f9058dd5a9e56cbf190d7b4515dddaaec5b910 Mon Sep 17 00:00:00 2001 From: cschaefer26 Date: Wed, 8 Jul 2020 15:03:57 +0200 Subject: [PATCH] Updated notebook. --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 96e688a..66e702f 100644 --- a/README.md +++ b/README.md @@ -19,8 +19,8 @@ does not use any attention. Hence, the required memory grows linearly with text ## UPDATE (08.07.2020) -Model is now compatible with [MelGAN](https://github.com/seungwonpark/melgan). Check out out our [Colab Notebook](https://colab.research.google.com/github/as-ideas/ForwardTacotron/blob/master/notebooks/synthesize.ipynb) -. +Model is now compatible with the [MelGAN](https://github.com/seungwonpark/melgan) vocoder for faster-than-realtime synthesis on CPU. +Check out the [Colab Notebook](https://colab.research.google.com/github/as-ideas/ForwardTacotron/blob/master/notebooks/synthesize.ipynb) where you can synthesize text using a pretrained MelGAN or WaveRNN model trained on LJSpeech. ## 🔈 Samples