Adapt the audio model to the new interface

lightx2v/models/runners/wan/wan_audio_runner.py

@@ -578,7 +578,7 @@ class WanAudioRunner(WanRunner):
         os.remove(out_path)
         os.remove(audio_file)
 
-    async def run_pipeline(self):
+    def run_pipeline(self):
         if self.config["use_prompt_enhancer"]:
             self.config["prompt_enhanced"] = self.post_prompt_enhancer()
 

scripts/wan/run_wan_i2v_audio.sh

@@ -40,7 +40,7 @@ python -m lightx2v.infer \
 --task i2v \
 --model_path $model_path \
 --config_json ${lightx2v_path}/configs/audio_driven/wan_i2v_audio.json \
---prompt_path ${lightx2v_path}/assets/inputs/audio/15.txt \
+--prompt  "The video features a old lady is saying something and knitting a sweater." \
 --negative_prompt 色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走 \
 --image_path ${lightx2v_path}/assets/inputs/audio/15.png \
 --audio_path ${lightx2v_path}/assets/inputs/audio/15.wav \