Skip to content

FunASR-powered video transcription, subtitle generation, and LLM-assisted clipping tool with a local Gradio UI.

bug fix

shixian.shi9b838f828e6d
diff --git a/ClipVideo/clipvideo/gradio_service.py b/ClipVideo/clipvideo/gradio_service.py
index 0cb8e06..84cba0b 100644
--- a/ClipVideo/clipvideo/gradio_service.py
+++ b/ClipVideo/clipvideo/gradio_service.py
@@ -81,7 +81,11 @@ if __name__ == "__main__":
'https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ClipVideo/2022%E4%BA%91%E6%A0%96%E5%A4%A7%E4%BC%9A_%E7%89%87%E6%AE%B5.mp4',
'https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ClipVideo/2022%E4%BA%91%E6%A0%96%E5%A4%A7%E4%BC%9A_%E7%89%87%E6%AE%B52.mp4',
'https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ClipVideo/%E4%BD%BF%E7%94%A8chatgpt_%E7%89%87%E6%AE%B5.mp4'],
- [video_input])
+ [video_input],
+ label='语音识别示例 ASR Demo')
+ gr.Examples(['https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ClipVideo/%E8%AE%BF%E8%B0%88.mp4'],
+ [video_input],
+ label='说话人切分示例 SD Demo')
video_sd_switch = gr.Radio(["no", "yes"], label="👥是否区分说话人 Recognize Speakers", value='no')
recog_button2 = gr.Button("👂识别 Recognize")
video_text_output = gr.Textbox(label="✏️识别结果 Recognition Result")
@@ -117,7 +121,7 @@ if __name__ == "__main__":
audio_spk_input = gr.Textbox(label="✏️待裁剪说话人 Speaker to Clip (多个说话人使用'#'连接)")
with gr.Row():
audio_start_ost = gr.Slider(minimum=-500, maximum=1000, value=0, step=50, label="⏪开始位置偏移 Start Offset (ms)")
- audio_end_ost = gr.Slider(minimum=-500, maximum=1000, value=100, step=50, label="⏩结束位置偏移 End Offset (ms)")
+ audio_end_ost = gr.Slider(minimum=-500, maximum=1000, value=0, step=50, label="⏩结束位置偏移 End Offset (ms)")
with gr.Row():
clip_button1 = gr.Button("✂️裁剪 Clip")
audio_output = gr.Audio(label="🔊裁剪结果 Audio Clipped")
diff --git a/ClipVideo/clipvideo/videoclipper.py b/ClipVideo/clipvideo/videoclipper.py
index a649076..66cc79c 100644
--- a/ClipVideo/clipvideo/videoclipper.py
+++ b/ClipVideo/clipvideo/videoclipper.py
@@ -67,7 +67,7 @@ class VideoClipper():
ts = proc_spk(_dest_spk, state['sd_sentences'])
for _ts in ts: all_ts.append(_ts)
ts = all_ts
- ts.sort()
+ # ts.sort()
srt_index = 0
clip_srt = ""
if len(ts):
@@ -130,7 +130,7 @@ class VideoClipper():
for _ts in ts: all_ts.append(_ts)
time_acc_ost = 0.0
ts = all_ts
- ts.sort()
+ # ts.sort()
clip_srt = ""
if len(ts):
start, end = ts[0][0] / 16000, ts[0][1] / 16000
@@ -148,14 +148,19 @@ class VideoClipper():
for _ts in ts[1:]:
start, end = _ts[0] / 16000, _ts[1] / 16000
srt_clip, subs, srt_index = generate_srt_clip(sentences, start, end, begin_index=srt_index-1, time_acc_ost=time_acc_ost)
+ chi_subs = []
+ sub_starts = subs[0][0][0]
+ for sub in subs:
+ chi_subs.append(((sub[0][0]-sub_starts, sub[0][1]-sub_starts), sub[1]))
start, end = start+start_ost/1000.0, end+end_ost/1000.0
_video_clip = video.subclip(start, end)
start_end_info += ", from {} to {}".format(start, end)
clip_srt += srt_clip
if add_sub:
generator = lambda txt: TextClip(txt, font='./font/STHeitiMedium.ttc', fontsize=font_size, color=font_color)
- subtitles = SubtitlesClip(subs, generator)
+ subtitles = SubtitlesClip(chi_subs, generator)
_video_clip = CompositeVideoClip([_video_clip, subtitles.set_pos(('center','bottom'))])
+ # _video_clip.write_videofile("debug.mp4", audio_codec="aac")
concate_clip.append(copy.copy(_video_clip))
time_acc_ost += end+end_ost/1000.0 - (start+start_ost/1000.0)
message = "{} periods found in the audio: ".format(len(ts)) + start_end_info