From e3e1ab331925b275d2ce327ade3ad87779d08e75 Mon Sep 17 00:00:00 2001 From: jinzr Date: Thu, 29 Feb 2024 10:45:03 +0800 Subject: [PATCH] Update README.md --- egs/ljspeech/TTS/README.md | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) diff --git a/egs/ljspeech/TTS/README.md b/egs/ljspeech/TTS/README.md index d3c734862..930826971 100644 --- a/egs/ljspeech/TTS/README.md +++ b/egs/ljspeech/TTS/README.md @@ -13,4 +13,24 @@ The above information is from the [LJSpeech website](https://keithito.com/LJ-Spe This recipe provides a VITS model trained on the LJSpeech dataset. -Pretrained model can be found [here](https://huggingface.co/Zengwei/icefall-tts-ljspeech-vits-2023-11-29). +Pretrained model can be found [here](https://huggingface.co/Zengwei/icefall-tts-ljspeech-vits-2024-02-28). + +The training command is given below: +``` +export CUDA_VISIBLE_DEVICES=0,1,2,3 +./vits/train.py \ + --world-size 4 \ + --num-epochs 1000 \ + --start-epoch 1 \ + --use-fp16 1 \ + --exp-dir vits/exp \ + --max-duration 500 +``` + +To inference, use: +``` +./vits/infer.py \ + --exp-dir vits/exp \ + --epoch 1000 \ + --tokens data/tokens.txt +``` \ No newline at end of file