From 2482b112c02cfc003ef267ac57b91132d2d8f678 Mon Sep 17 00:00:00 2001
From: TianYuan <white-sky@qq.com>
Date: Wed, 18 Aug 2021 03:45:58 +0000
Subject: [PATCH] fix readme

---
 examples/fastspeech2/baker/README.md | 1 +
 1 file changed, 1 insertion(+)

diff --git a/examples/fastspeech2/baker/README.md b/examples/fastspeech2/baker/README.md
index dac6a40..32bd109 100644
--- a/examples/fastspeech2/baker/README.md
+++ b/examples/fastspeech2/baker/README.md
@@ -22,6 +22,7 @@ Run the command below to preprocess the dataset.
 ```bash
 ./run.sh
 ```
+If you want to train fastspeech2 with cpu, please add `--device=cpu` arguments for `python3 train.py` in `run.sh`.
 ## Synthesize
 We use [parallel wavegan](https://github.com/PaddlePaddle/Parakeet/tree/develop/examples/parallelwave_gan/baker) as the neural vocoder.
 Download pretrained parallel wavegan model from [parallel_wavegan_baker_ckpt_0.4.zip](https://paddlespeech.bj.bcebos.com/Parakeet/parallel_wavegan_baker_ckpt_0.4.zip) and unzip it.