FunASR-powered video transcription, subtitle generation, and LLM-assisted clipping tool with a local Gradio UI.
Python ai-toolsai-video-editingasrauto-subtitleschinesecontent-creationfunasrfunclipgradiollmparaformerspeech-recognitionspeech-to-textsubtitles-generatortranscriptionvideo-editingvideo-processingvideo-subtitlesvideo-transcriptionwhisper-alternative
bug fix
维石212be946e84e
diff --git a/funclip/launch.py b/funclip/launch.pyindex 4209bca..077e299 100644--- a/funclip/launch.py+++ b/funclip/launch.py@@ -132,12 +132,30 @@ if __name__ == "__main__":if video_state is not None:clip_video_file, message, clip_srt = audio_clipper.video_clip(dest_text, start_ost, end_ost, video_state,- dest_spk=video_spk_input, output_dir=output_dir, timestamp_list=timestamp_list)+ dest_spk=video_spk_input, output_dir=output_dir, timestamp_list=timestamp_list, add_sub=False)return clip_video_file, None, message, clip_srtif audio_state is not None:(sr, res_audio), message, clip_srt = audio_clipper.clip(dest_text, start_ost, end_ost, audio_state,- dest_spk=video_spk_input, output_dir=output_dir, timestamp_list=timestamp_list)+ dest_spk=video_spk_input, output_dir=output_dir, timestamp_list=timestamp_list, add_sub=False)+ return None, (sr, res_audio), message, clip_srt++ def AI_clip_subti(LLM_res, dest_text, video_spk_input, start_ost, end_ost, video_state, audio_state, output_dir):+ timestamp_list = extract_timestamps(LLM_res)+ output_dir = output_dir.strip()+ if not len(output_dir):+ output_dir = None+ else:+ output_dir = os.path.abspath(output_dir)+ if video_state is not None:+ clip_video_file, message, clip_srt = audio_clipper.video_clip(+ dest_text, start_ost, end_ost, video_state,+ dest_spk=video_spk_input, output_dir=output_dir, timestamp_list=timestamp_list, add_sub=True)+ return clip_video_file, None, message, clip_srt+ if audio_state is not None:+ (sr, res_audio), message, clip_srt = audio_clipper.clip(+ dest_text, start_ost, end_ost, audio_state,+ dest_spk=video_spk_input, output_dir=output_dir, timestamp_list=timestamp_list, add_sub=True)return None, (sr, res_audio), message, clip_srt# gradio interface@@ -263,7 +281,18 @@ if __name__ == "__main__":video_end_ost,video_state,audio_state,- output_dir+ output_dir,+ ],+ outputs=[video_output, audio_output, clip_message, srt_clipped])+ llm_clip_subti_button.click(AI_clip_subti,+ inputs=[llm_result,+ video_text_input,+ video_spk_input,+ video_start_ost,+ video_end_ost,+ video_state,+ audio_state,+ output_dir,],outputs=[video_output, audio_output, clip_message, srt_clipped])diff --git a/funclip/utils/subtitle_utils.py b/funclip/utils/subtitle_utils.pyindex c90cce6..0f5ad92 100644--- a/funclip/utils/subtitle_utils.py+++ b/funclip/utils/subtitle_utils.py@@ -69,7 +69,8 @@ def generate_srt_clip(sentence_list, start, end, begin_index=0, time_acc_ost=0.0cc = 1 + begin_indexsubs = []for _, sent in enumerate(sentence_list):- sent['text'] = str2list(sent['text'])+ if isinstance(sent['text'], str):+ sent['text'] = str2list(sent['text'])if sent['timestamp'][-1][1] <= start:# print("CASE0")continue