-
#7785 [openrouter] Cannot connect to https://openrouter.ai/api/v1. Check network access or verify local server deployment.Connection error.:Traceback (most recent call last):...
💬 1[Last reply:1 month ago] The error WinError 10061 (connection actively refused) with the traceback showing httpcore\_sync\http_proxy.py indicates that the connection to https://openrouter.ai/api/v1 is being routed through a p...187.40**7 Posted at: 1 month ago -
#7784 Network connection failed:Traceback (most recent call last): File...
💬 1[Last reply:1 month ago] This is a network connectivity failure when calling the Gemini translation API. The error [WinError 10053] An established connection was aborted by the software in your host machine indicates the loca...187.40**7 Posted at: 1 month ago -
#7783 高级选项里,我主要翻译韩文视频,怎么设置正确率最高...
💬 1[Last reply:1 month ago] 翻译韩文视频时,提升“正确率”的关键不是单一参数,而是语音识别渠道/模型选择 + 字幕切分参数 + 翻译渠道三者配合。按以下设置最稳妥:1. 语音识别渠道(最关键)首选:faster-whisper(内置),模型选 large-v3(准确度最高)或 large-v3-turbo(速度与质量兼顾)备选:openai-whisper(内置) + large-v3若有 N 卡,务必勾选启用CUDA加速2...103.151**5 Posted at: 1 month ago -
#7782 ...
💬 1[Last reply:1 month ago] 这是模型文件下载不完整或缺失导致的。根据你提供的文件目录截图,你正在使用 F5-TTS(内置) 配音渠道,但 models/models--SWivid--F5-TTS/F5TTS_v1_Base 目录下的模型文件下载失败或未下载完全。解决方案:手动下载模型访问官方模型下载地址:https://huggingface.co/SWivid/F5-TTS/tree/main/F5TTS_v1_Base...103.151**5 Posted at: 1 month ago -
#7781 No audio file was generated during the listening test.self=OmniVoice(uuid=None, proxy_str=None, last_down_time=0, tts_type=3, len=1, language='de', play=True, volume='+33%', rate='...
💬 1[Last reply:1 month ago] The OmniVoice TTS channel failed to produce any output audio file during the listening test — meaning the model either did not load correctly or the voice-cloning reference was rejected, so no wav was...42.1**0 Posted at: 1 month ago -
#7772 下的有模型models--Qwen--Qwen3-ASR-1.7B 为什么还要再新建一个文件夹models--Qwen--Qwen3-ASR-1.7B-hf 再去下模型 怎么指定...
💬 8[Last reply:1 month ago] 4.12后改用qwen的hf版模型,所以需要重新下载,不能简单修改文件夹名称替换。Qwen3-ForcedAligner-0.6B 是在识别后重新断句用的模型,同样都改为了hf版hf版是qwen针对transformers推理的单独版本...140.207**4 Posted at: 1 month ago -
#7780 翻译字幕阶段出错 [Gemini AI] User location is not supported for the API use.Traceback (most recent call last):...
💬 1[Last reply:1 month ago] 核心原因:Gemini AI 接口不支持你当前所在的 IP 地区(User location is not supported for the API use),即节点地区不被 Google 允许。解决方案:更换网络代理节点,切换到 Gemini 支持的地区(如美国、日本等),并确保代理全局生效、端口(你填的 http://127.0.0.1:7897)正确。若换节点后仍不行,改用其他无需地区限制...103.151**5 Posted at: 1 month ago -
#7779 Connection reset, network may be unstableTraceback (most recent call last):...
💬 1[Last reply:1 month ago] 这是网络连接问题,微软免费翻译接口在请求时被服务器强制断开(WinError 10054)。通常由网络不稳定、代理配置异常、或微软接口临时限流导致。解决方案(按顺序尝试)检查并清空代理设置:查看主界面「对齐控制」行右侧的代理文本框。如果你填写了代理地址(如 http://127.0.0.1:10808),请临时清空它再重试。填写了无效或不通的代理会导致所有网络请求失败。更换翻译渠道:微软免费翻译接...101.71**1 Posted at: 1 month ago -
#7778 [Clone-Voice] Service must be deployed and running before useself.api_url='http://127.0.0.1:9988'...
💬 1[Last reply:1 month ago] The core cause: you selected the clone-voice(LocalAPI) TTS channel, but that local service is not running (or not deployed). The error WinError 10061 No connection could be made because the target mac...175.100**4 Posted at: 1 month ago -
#7777 Unsupported language_id 'km'. Supported languages: ar, da, de, el, en, es, fi, fr, he, hi, it, ja, ko, ms, nl, no, pl, pt, ru, sv, sw, tr, zhTraceback (most recent call last):...
💬 1[Last reply:1 month ago] The root cause is that you selected the ChatterBox (Built-in) TTS channel for dubbing, but ChatterBox does not support Khmer (km). Its supported languages are only: ar, da, de, el, en, es, fi, fr, he,...175.100**4 Posted at: 1 month ago -
#7776 该模型始终思考,不支持关闭思考;请使用 low、high 或 max。https://open.bigmodel.cn/api/paas/v4/...
💬 1[Last reply:1 month ago] 这是智谱AI渠道(GLM-5.3-Flash)强制思考模式导致的报错:该模型不支持关闭思考(软件内部会发送关闭思考的请求参数),必须显式指定思考强度为 low、high 或 max。解决方案:更换模型:在软件 菜单 → 翻译设置 → 智谱AI 中,将模型名从 GLM-5.3-Flash 改为不带强制思考的普通模型(如 glm-4-flash 等)。这类支持思考强度的新模型目前未被软件适配请求参数,...2602**84e Posted at: 1 month ago -
#7775 该模型始终思考,不支持关闭思考;请使用 low、high 或 max。https://open.bigmodel.cn/api/paas/v4/...
💬 1[Last reply:1 month ago] 该错误是智谱 API 返回的 400 报错:GLM-5.3-Flash 模型强制要求开启思考模式,且仅接受 low、high、max 三个档位,而 pyVideoTrans 尝试关闭思考模式(thinking 传了禁用值),导致参数被拒。解决方案(按顺序尝试):取消勾选 Thinking 复选框打开 菜单 → 翻译设置 → 智谱AI(GLM)设置窗口,将 Thinking(思考模式)复选框取消选中...2602**84e Posted at: 1 month ago -
#7774 Video file selection required...
💬 1[Last reply:1 month ago] "Video file selection required" means you clicked Start without selecting any input file. pyVideoTrans cannot begin any task until at least one audio/video file (or a folder) is chosen.Steps to fix:Cl...175.100**4 Posted at: 1 month ago -
#7773 配音语句断断续续的,不连贯,分句不好...
💬 1[Last reply:1 month ago] 配音断断续续、不连贯,核心原因是语音识别阶段VAD切分颗粒度过碎,导致每句字幕太短,配音逐句生成后再拼接,句与句之间必然出现停顿感。此外翻译渠道合并/拆分字幕也会加剧该问题。解决步骤:调整VAD切分参数(菜单 → 工具/选项 → 高级选项 → 语音识别参数):最长语音持续(秒):调大至 8~10最短语音持续(毫秒):调大至 3500~4000(防碎裂)静音分割持续(毫秒):调大至 800(停顿超过...39.68**1 Posted at: 1 month ago -
#7771 Authentication Fails, Your api key: ****c4a0 is invalid (request_id: a82b6ced-a11a-405d-bbc5-78447d0c12d3)https://api.deepseek.com/v1/...
💬 1[Last reply:1 month ago] 错误核心原因:DeepSeek API 密钥无效(api key: c4a0)。服务端返回 401 认证失败,与软件本身无关,属于密钥配置问题。常见原因:密钥复制错误(多复制了空格、换行,或复制不全)。密钥已在 DeepSeek 平台被删除或重置。账户未充值/余额不足导致密钥被停用。填错了位置(例如把别家平台的 key 填到了 DeepSeek 位置)。操作步骤:登录 DeepSeek 开放平台 ...39.68**1 Posted at: 1 month ago -
#7770 v4.14优化 语音/字幕/画面 对齐算法 有具体那些吗...
💬 3[Last reply:1 month ago] 知识库中无相关记录,无法确认 v4.14 是否修复了音频加速对齐时的机械音问题。关于音频加速的实现,软件采用两种方案(来源于知识库,但未提及是否针对机械音做过专门修复):首选 Rubber Band 算法(pyrubberband),效果最平滑;回退方案为 FFmpeg atempo 滤镜链。如需排查此问题,建议提供:软件版本号、是否安装 rubberband、以及日志文件底部约 30 行报错或处...2a14**8dc Posted at: 1 month ago -
#7769 [DEBUG] 测试Google翻译测试返回status_code=0[WARNING] 检测google翻译失败:status_code=0,改为使用微软翻译[DEBUG] 以纯文本行形式翻译,每次翻译10行,翻译后暂停0.0s[DEBU......
💬 3[Last reply:1 month ago] 这是因为 Qwen3-TTS 内置渠道在 macOS 上尝试使用 MPS(Apple 芯片 GPU 加速)加载模型,但 Apple MPS 后端不支持 BFloat16 数据类型,导致模型加载失败。故障定位:报错 TypeError: BFloat16 is not supported on MPS,发生在 qwen_tts 模型 from_pretrained 阶段,即模型下载后加载到 MPS ...194.56**4 Posted at: 1 month ago -
#7768 Video file selection required...
💬 1[Last reply:1 month ago] Core cause: no input file was selected. pyVideoTrans validates the input first and aborts the task with "Video file selection required" when the file list is empty.Solution:Click Select Video/Audio at...2803**8a1d Posted at: 1 month ago -
#7767 Multiple connection retries failed, service may be temporarily unavailableTraceback (most recent call last):...
💬 1[Last reply:1 month ago] This is a network DNS resolution failure. The error socket.gaierror: [Errno 11001] getaddrinfo failed on edge.microsoft.com means the Microsoft Translate (Free) channel cannot be reached — either your...177.91**2 Posted at: 1 month ago -
#7766 pyVideoTrans 转录翻译 视频日语语音翻译成字幕 发送完整字幕 选项 什么时候开启...
💬 1[Last reply:1 month ago] 发送完整字幕 选项在 AI 翻译渠道 下的作用,是决定翻译时发给 AI 模型的是 带时间轴的 SRT 格式内容,还是 纯逐行文本。结论:优先开启当字幕翻译渠道选用 AI 大模型(如 DeepSeek / OpenAI ChatGPT / Gemini 等)时,建议开启 发送完整字幕。它能让 AI 看到字幕的行号、时间轴和上下文,翻译更连贯、断句更合理,是提升翻译质量的关键设置之一。对于日语视频翻译...111.31**4 Posted at: 1 month ago
Open source and free maintenance is not easy. If this project is helpful to you, please consider making a small donation to help the project continue to maintain and update.