diff --git a/doc/doc_ch/recognition.md b/doc/doc_ch/recognition.md
index bb7d01712a85c92a02109e41814059e6c98c7cdc..4019aa123f6bf1ab0e890b68fbbcd008eb86e440 100644
--- a/doc/doc_ch/recognition.md
+++ b/doc/doc_ch/recognition.md
@@ -235,7 +235,10 @@ PaddleOCR支持训练和评估交替进行, 可以在 `configs/rec/rec_icdar15_t
 | rec_r31_sar.yml               | SAR | ResNet31 | None | LSTM encoder | LSTM decoder |
 | rec_resnet_stn_bilstm_att.yml | SEED | Aster_Resnet | STN | BiLSTM | att |
 
-*其中SEED模型需要额外加载FastText训练好的[语言模型](https://dl.fbaipublicfiles.com/fasttext/vectors-crawl/cc.en.300.bin.gz)
+*其中SEED模型需要额外加载FastText训练好的[语言模型](https://dl.fbaipublicfiles.com/fasttext/vectors-crawl/cc.en.300.bin.gz) ,并且安装 fasttext 依赖：
+```
+python3.7 -m pip install fasttext==0.9.1
+```
 
 训练中文数据，推荐使用[rec_chinese_lite_train_v2.0.yml](../../configs/rec/ch_ppocr_v2.0/rec_chinese_lite_train_v2.0.yml)，如您希望尝试其他算法在中文数据集上的效果，请参考下列说明修改配置文件：
 
diff --git a/ppocr/data/imaug/operators.py b/ppocr/data/imaug/operators.py
index c3dfd316f86d88b5c7fd52eb6ae23d22a4dd32eb..daa67a25dae93dde74fc0b92aad4aa6ef4d4c003 100644
--- a/ppocr/data/imaug/operators.py
+++ b/ppocr/data/imaug/operators.py
@@ -23,7 +23,6 @@ import sys
 import six
 import cv2
 import numpy as np
-import fasttext
 
 
 class DecodeImage(object):
@@ -136,6 +135,7 @@ class ToCHWImage(object):
 
 class Fasttext(object):
     def __init__(self, path="None", **kwargs):
+        import fasttext
         self.fast_model = fasttext.load_model(path)
 
     def __call__(self, data):
diff --git a/requirements.txt b/requirements.txt
index 9900588b25df99e0853ec4521f0632578c55f530..5e999790397c1530f251993e9f5b5d0f2d67b753 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -12,5 +12,4 @@ cython
 lxml
 premailer
 openpyxl
-fasttext==0.9.1