From 2a0dd5447adac3a917516dfb9320a4a3494c77e7 Mon Sep 17 00:00:00 2001
From: "huzetao.hzt" <huzetao.hzt@alibaba-inc.com>
Date: Tue, 7 Jan 2025 17:29:20 +0800
Subject: [PATCH] add usage

---
 README.md | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/README.md b/README.md
index d5fc9ea..8ccca02 100644
--- a/README.md
+++ b/README.md
@@ -128,6 +128,8 @@ import torchaudio
 
 **CosyVoice2 Usage**
 ```python
+# NOTE if you want to use tensorRT to accerlate the flow matching inference, please set load_trt=True.
+# if you don't want to save tensorRT model on disk, please set environment variable `NOT_SAVE_TRT=1`.
 cosyvoice = CosyVoice2('pretrained_models/CosyVoice2-0.5B', load_jit=False, load_trt=False, fp16=False)
 
 # NOTE if you want to reproduce the results on https://funaudiollm.github.io/cosyvoice2, please add text_frontend=False during inference