Open leafopening opened 2 weeks ago
我在32G显存的v100下跑实时推理示例:python -m scripts.realtime_inference --inference_config configs/inference/realtime.yaml --batch_size 4
得到的结果是“Total process time of 202 frames including saving images = 25.92679262161255s ”。
配置文件是这样的: avator_1: preparation: False bbox_shift: 5 video_path: "data/video/yongen.mp4" audio_clips: audio_0: "data/audio/yongen.wav"
请问如何调整??
https://github.com/TMElyralab/MuseTalk/issues/49#issuecomment-2081429256 这里有我们的测速结果,你可以对比一下
我在32G显存的v100下跑实时推理示例:python -m scripts.realtime_inference --inference_config configs/inference/realtime.yaml --batch_size 4
得到的结果是“Total process time of 202 frames including saving images = 25.92679262161255s ”。
配置文件是这样的: avator_1: preparation: False bbox_shift: 5 video_path: "data/video/yongen.mp4" audio_clips: audio_0: "data/audio/yongen.wav"
请问如何调整??