Support sekotalk multiperson (#321)

Co-authored-by: PengGao <peng.gaoc@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Yang Yong (雍洋) <yongyang1030@163.com>

scripts/seko_talk/run_seko_talk_13_fp8_dist_bucket_shape_8gpus_5s_realtime.sh

@@ -20,5 +20,5 @@ torchrun --nproc-per-node 8 -m lightx2v.infer \
 --prompt  "The video features a male speaking to the camera with arms spread out, a slightly furrowed brow, and a focused gaze." \
 --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/seko_input.png \
---audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.wav \
+--audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.mp3 \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_seko_talk.mp4

scripts/seko_talk/run_seko_talk_14_fp8_dist_bucket_shape_8gpus_1s_realtime.sh

@@ -20,5 +20,5 @@ torchrun --nproc-per-node 8 -m lightx2v.infer \
 --prompt  "The video features a male speaking to the camera with arms spread out, a slightly furrowed brow, and a focused gaze." \
 --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/seko_input.png \
---audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.wav \
+--audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.mp3 \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_seko_talk.mp4

scripts/seko_talk/run_seko_talk_15_base_compile.sh

@@ -20,5 +20,5 @@ python -m lightx2v.infer \
 --prompt  "The video features a male speaking to the camera with arms spread out, a slightly furrowed brow, and a focused gaze." \
 --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/seko_input.png \
---audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.wav \
+--audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.mp3 \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_seko_talk.mp4

scripts/seko_talk/run_seko_talk_16_fp8_dist_compile.sh

@@ -20,5 +20,5 @@ torchrun --nproc-per-node 8 -m lightx2v.infer \
 --prompt  "The video features a male speaking to the camera with arms spread out, a slightly furrowed brow, and a focused gaze." \
 --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/seko_input.png \
---audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.wav \
+--audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.mp3 \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_seko_talk.mp4

test_cases/run_seko_talk_01_base.sh

@@ -20,5 +20,5 @@ python -m lightx2v.infer \
 --prompt  "The video features a male speaking to the camera with arms spread out, a slightly furrowed brow, and a focused gaze." \
 --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/seko_input.png \
---audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.wav \
+--audio_path ${lightx2v_path}/assets/inputs/audio/seko_input.mp3 \
 --save_video_path ${lightx2v_path}/save_results/output_lightx2v_seko_talk.mp4