Update README.md

缺少glm speech_tokenizer
Traceback (most recent call last):
File "/home/zhibo/workspace/Kimi-Audio/test.py", line 3, in <module>
from kimia_infer.api.kimia import KimiAudio
File "/home/zhibo/workspace/Kimi-Audio/kimia_infer/api/kimia.py", line 10, in <module>
from .prompt_manager import KimiAPromptManager
File "/home/zhibo/workspace/Kimi-Audio/kimia_infer/api/prompt_manager.py", line 12, in <module>
from kimia_infer.models.tokenizer.glm4_tokenizer import Glm4Tokenizer
File "/home/zhibo/workspace/Kimi-Audio/kimia_infer/models/tokenizer/glm4_tokenizer.py", line 6, in <module>
from .glm4.speech_tokenizer.modeling_whisper import WhisperVQEncoder
ModuleNotFoundError: No module named 'kimia_infer.models.tokenizer.glm4.speech_tokenizer'

README.md

@@ -44,6 +44,7 @@ For more details, please refer to our [GitHub Repository](https://github.com/Moo
 We recommend that you build a Docker image to run the inference. After cloning the inference code, you can construct the image using the `docker build` command.
 ```bash
 git clone https://github.com/MoonshotAI/Kimi-Audio
+git submodule update --init
 cd Kimi-Audio
 docker build -t kimi-audio:v0.1 .
 ```