From 1da23aa25978a2aa36a6dda66bde50b3d4819eec Mon Sep 17 00:00:00 2001 From: Downupanddownup Date: Thu, 25 Apr 2024 22:54:40 +0800 Subject: [PATCH] =?UTF-8?q?bug=E4=BF=AE=E5=A4=8D?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .../{config => config_param}/__init__.py | 0 .../config_manager.py | 0 .../{config => config_param}/config_params.py | 2 +- .../ref_audio_selector_webui.py | 26 +++++++++---------- .../tool/asr/funasr_asr_multi_level_dir.py | 2 +- Ref_Audio_Selector/tool/audio_inference.py | 2 +- .../tool/delete_inference_with_ref.py | 6 ++--- .../tool/text_comparison/asr_text_process.py | 4 +-- 8 files changed, 21 insertions(+), 21 deletions(-) rename Ref_Audio_Selector/{config => config_param}/__init__.py (100%) rename Ref_Audio_Selector/{config => config_param}/config_manager.py (100%) rename Ref_Audio_Selector/{config => config_param}/config_params.py (96%) diff --git a/Ref_Audio_Selector/config/__init__.py b/Ref_Audio_Selector/config_param/__init__.py similarity index 100% rename from Ref_Audio_Selector/config/__init__.py rename to Ref_Audio_Selector/config_param/__init__.py diff --git a/Ref_Audio_Selector/config/config_manager.py b/Ref_Audio_Selector/config_param/config_manager.py similarity index 100% rename from Ref_Audio_Selector/config/config_manager.py rename to Ref_Audio_Selector/config_param/config_manager.py diff --git a/Ref_Audio_Selector/config/config_params.py b/Ref_Audio_Selector/config_param/config_params.py similarity index 96% rename from Ref_Audio_Selector/config/config_params.py rename to Ref_Audio_Selector/config_param/config_params.py index 74c77c6..015a5e7 100644 --- a/Ref_Audio_Selector/config/config_params.py +++ b/Ref_Audio_Selector/config_param/config_params.py @@ -1,4 +1,4 @@ -import Ref_Audio_Selector.config.config_manager as config_manager +import Ref_Audio_Selector.config_param.config_manager as config_manager config = config_manager.get_config() diff --git a/Ref_Audio_Selector/ref_audio_selector_webui.py b/Ref_Audio_Selector/ref_audio_selector_webui.py index 40eb9d8..710aa3c 100644 --- a/Ref_Audio_Selector/ref_audio_selector_webui.py +++ b/Ref_Audio_Selector/ref_audio_selector_webui.py @@ -8,7 +8,7 @@ import Ref_Audio_Selector.tool.audio_inference as audio_inference import Ref_Audio_Selector.tool.audio_config as audio_config import Ref_Audio_Selector.tool.delete_inference_with_ref as delete_inference_with_ref import Ref_Audio_Selector.common.common as common -import Ref_Audio_Selector.config.config_params as params +import Ref_Audio_Selector.config_param.config_params as params from tools.i18n.i18n import I18nAuto from config import python_exec, is_half from tools import my_utils @@ -176,7 +176,7 @@ def model_inference(text_work_space_dir, text_role, text_model_inference_voice_d traceback.print_exc() text_model_inference_info = f"发生异常：{e}" text_asr_audio_dir = '' - return i18n(text_model_inference_info), text_asr_audio_dir + return i18n(text_model_inference_info), text_asr_audio_dir, text_asr_audio_dir # 对推理生成音频执行asr @@ -371,9 +371,9 @@ def save_work_dir(text_work_space_dir, text_role): else: role_dir = '' for i in range(1, 101): - dir_name = os.path.join(text_work_space_dir, f"role_{i}") + role_dir = f"role_{i}" + dir_name = os.path.join(text_work_space_dir, role_dir) if not os.path.isdir(dir_name): - role_dir = dir_name break rw_param.write_role(role_dir) return role_dir @@ -398,7 +398,7 @@ with gr.Blocks() as app: button_convert_from_list = gr.Button(i18n("开始生成待参考列表"), variant="primary") text_convert_from_list_info = gr.Text(label=i18n("参考列表生成结果"), value="", interactive=False) gr.Markdown(value=i18n("1.2：选择基准音频，执行相似度匹配，并分段随机抽样")) - text_sample_dir = gr.Text(label=i18n("参考音频抽样目录"), value="", interactive=False) + text_sample_dir = gr.Text(label=i18n("参考音频抽样目录"), value="", interactive=True) button_convert_from_list.click(convert_from_list, [text_work_space_dir, text_role, text_list_input], [text_convert_from_list_info, text_sample_dir]) with gr.Row(): @@ -440,11 +440,7 @@ with gr.Blocks() as app: text_model_inference_info = gr.Text(label=i18n("批量推理结果"), value="", interactive=False) with gr.Accordion(label=i18n("第三步：进行参考音频效果校验与筛选"), open=False): gr.Markdown(value=i18n("3.1：启动asr，获取推理音频文本")) - text_asr_audio_dir = gr.Text(label=i18n("待asr的音频所在目录"), value="", interactive=False) - button_model_inference.click(model_inference, - [text_work_space_dir, text_role, text_model_inference_voice_dir, text_url, - text_text, text_ref_path, text_ref_text, text_emotion, - text_test_content], [text_model_inference_info, text_asr_audio_dir]) + text_asr_audio_dir = gr.Text(label=i18n("待asr的音频所在目录"), value="", interactive=True) with gr.Row(): dropdown_asr_model = gr.Dropdown( label=i18n("ASR 模型"), @@ -468,7 +464,7 @@ with gr.Blocks() as app: button_asr = gr.Button(i18n("启动asr"), variant="primary") text_asr_info = gr.Text(label=i18n("asr结果"), value="", interactive=False) gr.Markdown(value=i18n("3.2：启动文本相似度分析")) - text_text_similarity_analysis_path = gr.Text(label=i18n("待分析的文件路径"), value="", interactive=False) + text_text_similarity_analysis_path = gr.Text(label=i18n("待分析的文件路径"), value="", interactive=True) button_asr.click(asr, [text_work_space_dir, text_role, text_asr_audio_dir, dropdown_asr_model, dropdown_asr_size, dropdown_asr_lang], [text_asr_info, text_text_similarity_analysis_path]) @@ -491,7 +487,7 @@ with gr.Blocks() as app: text_compare_audio_dir], [text_similarity_audio_output_info]) with gr.Row(): text_sync_ref_audio_dir = gr.Text(label=i18n("参考音频路径"), value="", interactive=True) - text_sync_inference_audio_dir = gr.Text(label=i18n("被同步的推理音频路径"), value="", interactive=False) + text_sync_inference_audio_dir = gr.Text(label=i18n("被同步的推理音频路径"), value="", interactive=True) with gr.Row(): button_sync_ref_audio = gr.Button(i18n("将参考音频的删除情况，同步到推理音频目录"), variant="primary") text_sync_ref_info = gr.Text(label=i18n("同步结果"), value="", interactive=False) @@ -501,7 +497,7 @@ with gr.Blocks() as app: gr.Markdown(value=i18n("4.1：编辑模板")) default_template_path = params.default_template_path default_template_content = common.read_file(default_template_path) - text_template_path = gr.Text(label=i18n("模板文件路径"), value=default_template_path, interactive=False) + text_template_path = gr.Text(label=i18n("模板文件路径"), value=default_template_path, interactive=True) text_template = gr.Text(label=i18n("模板内容"), value=default_template_content, lines=10) gr.Markdown(value=i18n("4.2：生成配置")) text_sync_ref_audio_dir2 = gr.Text(label=i18n("参考音频路径"), value="", interactive=True) @@ -515,6 +511,10 @@ with gr.Blocks() as app: text_subsection_num, text_sample_num, checkbox_similarity_output], [text_sample_info, text_model_inference_voice_dir, text_sync_ref_audio_dir, text_sync_ref_audio_dir2]) + button_model_inference.click(model_inference, + [text_work_space_dir, text_role, text_model_inference_voice_dir, text_url, + text_text, text_ref_path, text_ref_text, text_emotion, + text_test_content], [text_model_inference_info, text_asr_audio_dir, text_sync_inference_audio_dir]) app.launch( server_port=9423, diff --git a/Ref_Audio_Selector/tool/asr/funasr_asr_multi_level_dir.py b/Ref_Audio_Selector/tool/asr/funasr_asr_multi_level_dir.py index 3fbb021..d6e04c8 100644 --- a/Ref_Audio_Selector/tool/asr/funasr_asr_multi_level_dir.py +++ b/Ref_Audio_Selector/tool/asr/funasr_asr_multi_level_dir.py @@ -3,7 +3,7 @@ import argparse import os import traceback -import Ref_Audio_Selector.config.config_params as params +import Ref_Audio_Selector.config_param.config_params as params from tqdm import tqdm from funasr import AutoModel diff --git a/Ref_Audio_Selector/tool/audio_inference.py b/Ref_Audio_Selector/tool/audio_inference.py index efa4746..a93328d 100644 --- a/Ref_Audio_Selector/tool/audio_inference.py +++ b/Ref_Audio_Selector/tool/audio_inference.py @@ -1,7 +1,7 @@ import os import requests import itertools -import Ref_Audio_Selector.config.config_params as params +import Ref_Audio_Selector.config_param.config_params as params from urllib.parse import urlparse, parse_qs, urlencode, urlunparse, quote diff --git a/Ref_Audio_Selector/tool/delete_inference_with_ref.py b/Ref_Audio_Selector/tool/delete_inference_with_ref.py index 74ff9d9..4d7f175 100644 --- a/Ref_Audio_Selector/tool/delete_inference_with_ref.py +++ b/Ref_Audio_Selector/tool/delete_inference_with_ref.py @@ -1,7 +1,7 @@ import os import shutil import Ref_Audio_Selector.common.common as common -import Ref_Audio_Selector.config.config_params as params +import Ref_Audio_Selector.config_param.config_params as params def remove_matching_audio_files_in_text_dir(text_dir, emotions_list): @@ -11,7 +11,7 @@ def remove_matching_audio_files_in_text_dir(text_dir, emotions_list): emotion_tag = emotion_dict['emotion'] wav_file_name = f"{emotion_tag}.wav" file_path = os.path.join(root, wav_file_name) - if os.path.exists(file_path): + if not os.path.exists(file_path): print(f"Deleting file: {file_path}") try: os.remove(file_path) @@ -38,7 +38,7 @@ def delete_emotion_subdirectories(emotion_dir, emotions_list): folder_path = os.path.join(emotion_dir, emotion_folder) # 检查emotion子目录是否存在 - if os.path.isdir(folder_path): + if not os.path.isdir(folder_path): print(f"Deleting directory: {folder_path}") try: # 使用shutil.rmtree删除整个子目录及其内容 diff --git a/Ref_Audio_Selector/tool/text_comparison/asr_text_process.py b/Ref_Audio_Selector/tool/text_comparison/asr_text_process.py index 49b27ba..79e2f1f 100644 --- a/Ref_Audio_Selector/tool/text_comparison/asr_text_process.py +++ b/Ref_Audio_Selector/tool/text_comparison/asr_text_process.py @@ -3,7 +3,7 @@ import argparse from collections import defaultdict from operator import itemgetter import Ref_Audio_Selector.tool.text_comparison.text_comparison as text_comparison -import Ref_Audio_Selector.config.config_params as params +import Ref_Audio_Selector.config_param.config_params as params import Ref_Audio_Selector.common.common as common @@ -34,7 +34,7 @@ def parse_asr_file(file_path): def calculate_similarity_and_append_to_list(input_list, boundary): for item in input_list: - similarity_score = text_comparison.calculate_result(item['original_text'], item['asr_text'], boundary) + _, similarity_score = text_comparison.calculate_result(item['original_text'], item['asr_text'], boundary) item['similarity_score'] = similarity_score return input_list