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
update introduction
维石7f0a28a8538f
diff --git a/funclip/introduction.py b/funclip/introduction.pyindex ac731be..fa1827a 100644--- a/funclip/introduction.py+++ b/funclip/introduction.py@@ -1,22 +1,20 @@top_md_1 = ("""- 基于阿里巴巴通义实验室自研SeACo-Paraformer-长音频版的语音识别、端点检测、标点预测、时间戳功能、角色区分、热词定制化功能+ <div align="center">+ <div style="display:flex; gap: 0.25rem;" align="center">+ FunClip: <a href='https://github.com/alibaba-damo-academy/FunClip'><img src='https://img.shields.io/badge/Github-Code-blue'></a>+ 🌟支持我们: <a href='https://github.com/alibaba-damo-academy/FunClip/stargazers'><img src='https://img.shields.io/github/stars/alibaba-damo-academy/FunClip.svg?style=social'></a>+ </div>+ </div>++ 基于阿里巴巴通义实验室自研并开源的[FunASR](https://github.com/alibaba-damo-academy/FunASR)工具包及Paraformer系列模型及语音识别、端点检测、标点预测、时间戳预测、说话人区分、热词定制化开源链路准确识别,自由复制所需段落,或者设置说话人标识,一键裁剪、添加字幕* Step1: 上传视频或音频文件(或使用下方的用例体验),点击 **<font color="#f7802b">识别</font>** 按钮* Step2: 复制识别结果中所需的文字至右上方,或者右设置说话人标识,设置偏移与字幕配置(可选)* Step3: 点击 **<font color="#f7802b">裁剪</font>** 按钮或 **<font color="#f7802b">裁剪并添加字幕</font>** 按钮获得结果- """)--top_md_2 = ("""- 受到网络传输与服务资源的限制,用于体验的视频最好大小在40mb以下- 过大的视频可以尝试分离音轨使用音频剪辑,或 **<font color="#1785c4">通过源代码将您的ClipVideo服务部署在本地(推荐)</font>** :- <div align="center">- <div style="display:flex; gap: 0.25rem;" align="center">- FunASR_APP: <a href='https://github.com/alibaba/funasr-app'><img src='https://img.shields.io/badge/Github-Code-blue'></a>- 🌟支持我们: <a href='https://github.com/alibaba/funasr-app/stargazers'><img src='https://img.shields.io/github/stars/alibaba/funasr-app.svg?style=social'></a>- </div>- </div>++ 🔥 FunClip现在集成了大语言模型智能剪辑功能,选择LLM模型进行体验吧~""")top_md_3 = ("""访问FunASR项目与论文能够帮助您深入了解ParaClipper中所使用的语音处理相关模型:@@ -34,7 +32,6 @@ top_md_4 = ("""我们在「LLM智能裁剪」模块中提供三种LLM调用方2. 选择GPT开头的模型即为调用openai官方api,此时需要您自备sk与网络环境;3. [gpt4free](https://github.com/xtekky/gpt4free?tab=readme-ov-file)项目也被集成进FunClip,可以通过它免费调用gpt模型;- 其中方式1与方式2需要在界面中传入相应的apikey-+ 其中方式1与方式2需要在界面中传入相应的apikey方式3而可能非常不稳定,返回时间可能很长或者结果获取失败,可以多多尝试或者自己准备sk使用方式1""")diff --git a/funclip/launch.py b/funclip/launch.pyindex f89ba64..9813c10 100644--- a/funclip/launch.py+++ b/funclip/launch.py@@ -8,7 +8,7 @@ import loggingimport gradio as grfrom funasr import AutoModelfrom videoclipper import VideoClipper-from introduction import top_md_1, top_md_2, top_md_3, top_md_4+from introduction import top_md_1, top_md_3, top_md_4from llm.openai_api import openai_callfrom llm.g4f_openai_api import g4f_openai_callfrom llm.qwen_api import call_qwen_model@@ -124,7 +124,7 @@ if __name__ == "__main__":theme = gr.Theme.load("funclip/utils/theme.json")with gr.Blocks(theme=theme) as funclip_service:gr.Markdown(top_md_1)- gr.Markdown(top_md_2)+ # gr.Markdown(top_md_2)gr.Markdown(top_md_3)gr.Markdown(top_md_4)video_state, audio_state = gr.State(), gr.State()diff --git a/funclip/videoclipper.py b/funclip/videoclipper.pyindex 5e379b4..a6ff23a 100644--- a/funclip/videoclipper.py+++ b/funclip/videoclipper.py@@ -52,7 +52,8 @@ class VideoClipper():return_raw_text=True,is_final=True,hotword=hotwords,- output_dir=output_dir,)+ output_dir=output_dir,+ cache={})res_srt = generate_srt(rec_result[0]['sentence_info'])state['recog_res_raw'] = rec_result[0]['raw_text']state['timestamp'] = rec_result[0]['timestamp']