-
#7950 3ea9ce6ca9 split no voice error:FFmpeg 异常退出，但未提供具体错误。最近日志: Output #0, mp4, to 'novoice.mp4': Metadata: major_brand : isom minor_version : 512 compatible_brands: isomiso2avc1m...
💬 1[Last reply:8 minutes ago] 根据日志判断，失败的根因是：原始视频是 10bit 像素格式（yuv420p10le），而软件在“对齐”阶段调用 libx264（8bit 编码器）分离无声视频流时无法处理 10bit 输入，导致 FFmpeg 以 (returncode=1) 异常退出，抛出 split no voice error。日志中的关键证据：Video: h264 (avc1), yuv420p10le ... lib...206.14**3 Posted at: 8 minutes ago -
#7949 Network connection failed:{'text': 'Hello, my dear friend, I hope every day of yours is wonderful and joyful!', 'rate': '+0%', 'role': 'Yan(Female/HK)', 'filename': 'C:\\Users\\Mor...
💬 1[Last reply:2 hours ago] This is a 403 handshake rejection from Microsoft's free Edge-TTS endpoint (wss://speech.platform.bing.com/...). It is not a bug in pyVideoTrans — it means your request was blocked at the network/IP le...41.239**4 Posted at: 2 hours ago -
#7948 Secure connection failed, check system time or network settingsurllib3.exceptions.SSLError: EOF occurred in violation of protocol (_ssl.c:1017)...
💬 1[Last reply:2 hours ago] 这是网络代理导致的 SSL 握手失败，与软件本身无关。从日志看：你选了 Google(免费) 翻译，但因国内无法直连 Google，软件自动回退到微软翻译（edge.microsoft.com），而你在主界面代理文本框填了 http://127.0.0.1:7897，请求被强制走本地代理，代理节点握手时被切断（EOF），因此报 SSLError。解决方案（按优先级）方案一：清空代理（最快）主界面 ...2602**b8 Posted at: 2 hours ago -
#7947 如何开启上下文文感知不报错...
💬 1[Last reply:2 hours ago] 「启用上下文感知」会显著增加 GPU 显存占用，报错基本都是显存不足（CUDA out of memory / Unable to allocate）导致的。按以下顺序调优即可在开启它的同时避免报错：换小模型：把 faster-whisper 模型从 large-v3 降到 medium 或 small（large-v3 基础运行就需约 8GB 空闲显存）。降计算精度：菜单 → 工具/选项 → 高...59.36**9 Posted at: 2 hours ago -
#7946 The maximum decoding length must be > 0:Traceback (most recent call last): File...
💬 1[Last reply:2 hours ago] 该报错出现在 faster-whisper 内部解码阶段（generate_with_fallback → restore_speech_timestamps），与你勾选的「启用上下文感知(Enable context awareness)」直接相关。知识库中已明确说明：启用上下文感知会占用更多 GPU、更容易出现重复或幻觉，属于不稳定参数。处理步骤：打开 菜单(Menu) → 工具/选项(Too...59.36**9 Posted at: 2 hours ago -
#7945 翻译过程中出现 啊啊啊啊啊啊啊 啊啊啊啊啊啊啊啊 有几十个啊...
💬 1[Last reply:2 hours ago] 大量重复的「啊啊啊…」通常来自两个环节之一，需先确认出现在哪一步：① 若出现在「语音识别」生成的原字幕里（右侧字幕区先显示原文时就有）这是 ASR 幻觉（Whisper 系列在静音段、纯音乐段、拖长音处极易复读同一个字）。打开 菜单-工具/选项-高级选项-语音识别参数将 重复惩罚/(repetition penalty) 调大（如 1.3~1.5）将 文本压缩率/(compression rati...59.36**9 Posted at: 2 hours ago -
#7944 Error no file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory D:/win-pyvideotrans-v4.14/models/models--Qwen--Qwen...
💬 1[Last reply:3 hours ago] 这是模型下载不完整或下载失败导致的。Qwen3-TTS(内置) 的模型文件未成功写入本地目录，软件启动时找不到权重文件（model.safetensors 等），因此报错。此外，你选中了 clone 角色（音色克隆），Qwen3-TTS 的克隆功能依赖 Qwen3-TTS-12Hz-0.6B-Base 模型，该模型缺失时必然报此错。解决方法（任选其一）方法一：清空后重新自动下载（最简单）关闭软件；...2602**b8 Posted at: 3 hours ago -
#7943 CUDA error: device-side assert triggeredCompile with `TORCH_USE_CUDA_DSA` to enable device-side assertions....
💬 1[Last reply:3 hours ago] The CUDA error: device-side assert triggered is a GPU-side data validity error (in ChatterBox TTS this happens inside s3gen.embed_ref, i.e. the reference audio slice or token index fed to the GPU is o...2800**1911 Posted at: 3 hours ago -
#7942 Unable to open file 'model.bin' in model 'E://models/models--mobiuslabsgmbh--faster-whisper-large-v3-turbo',Traceback (most recent call last): File...
💬 1[Last reply:4 hours ago] The error means the faster-whisper model file model.bin does not exist in E:/models/models--mobiuslabsgmbh--faster-whisper-large-v3-turbo — i.e. the model was never downloaded completely (download fai...197.133**8 Posted at: 4 hours ago -
#7941 配音阶段出错 [clone-voice(本地API)] [Clone-Voice] 该渠道需另行部署并启动后可用self.api_url='http://127.0.0.1:9988'...
💬 1[Last reply:4 hours ago] 核心原因：clone-voice(本地API) 属于“本地自行部署”渠道，软件只是通过 http://127.0.0.1:9988 去调用你本机已启动的 clone-voice 服务；当前该端口没有任何服务在监听，因此被系统拒绝连接（WinError 10061）。软件本身不包含该模型，必须先单独部署并启动服务，否则必然报此错。解决方案：若坚持使用 clone-voice：需自行下载部署并启动 c...2602**b8 Posted at: 4 hours ago -
#7940 Secure connection failed, check system time or network settingsTraceback (most recent call last):...
💬 1[Last reply:5 hours ago] 这是访问微软免费翻译接口（edge.microsoft.com）时 TLS 握手失败，报错 SSLEOFError: EOF occurred in violation of protocol，并非软件本身缺陷，通常是网络环境或系统时间异常导致。常见原因及排查步骤：系统时间/时区错误（报错文本已提示）：右键任务栏时间 → 调整日期和时间 → 开启"自动设置时间"，或手动校准为正确时间后重试。代理冲...2409**b6b9 Posted at: 5 hours ago -
#7939 Network connection failedTraceback (most recent call last):...
💬 1[Last reply:6 hours ago] 核心原因：您选择的语音识别渠道是 Google识别API(免费)（videotrans/recognition/_google.py），该接口在国内网络下无法直连，WinError 10054 远程主机强迫关闭连接 即代理未生效或代理服务未运行导致连接被重置。解决方案（按顺序尝试）：首选：改用本地识别渠道，彻底摆脱该报错在 语音识别(STT Service) 下拉中选择 faster-whispe...2a14**dcf Posted at: 6 hours ago -
#7938 Video file selection required...
💬 1[Last reply:6 hours ago] This error means the software did not receive a valid input file path when you clicked Start — the "Select Video/Audio" field is still empty (or the path it holds is no longer valid/readable).Solution...87.199**2 Posted at: 6 hours ago -
#7937 Network connection failedTraceback (most recent call last):...
💬 1[Last reply:6 hours ago] 核心原因：你选择的语音识别渠道 Google识别API(免费) 需要访问 Google 服务器，而当前网络无法连接（WinError 10060 连接超时），即没有可用的科学上网环境或未在软件中填写代理。同时你的翻译渠道也是 Google(免费)，同样会失败。解决方案（二选一）：方案一（推荐，最简单）：改用本地识别渠道主界面 语音识别(STT Service) 渠道改为 faster-whispe...45.59**7 Posted at: 6 hours ago -
#7936 Dubbing failed completely:403, message='Invalid response status', url='wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1?TrustedClientToken=6A5AA1D4EAFF4E...
💬 1[Last reply:6 hours ago] Cause: HTTP 403 from wss://speech.platform.bing.com/.../edge/v1 means Microsoft's free Edge-TTS endpoint has temporarily rate-limited/blocked your outbound IP (or the proxy node you are routed through...90.1**0 Posted at: 6 hours ago -
#7935 Reference audio 'nu_viet.wav' not found in 'C:/Users/Admin/Downloads/win-pyvideotrans-v4.15/f5-tts' folder. Ensure file is WAV format and saved in target folder....
💬 1[Last reply:7 hours ago] The error indicates that the voice-cloning reference audio file nu_viet.wav is not present in the expected folder: C:/Users//Downloads/win-pyvideotrans-v4.15/f5-tts. Custom reference audio for cl...2001**b3b1 Posted at: 7 hours ago -
#7934 Dubbing failed completely:403, message='Invalid response status', url='wss://speech.platform.bing.com/consumer/speech/synthesize/readaloud/edge/v1?TrustedClientToken=6A5AA1D4EAFF4E...
💬 1[Last reply:7 hours ago] This is an Edge-TTS rate-limit block (HTTP 403), not a bug in your configuration. Edge-TTS uses Microsoft's free public WebSocket endpoint; when too many requests are sent in a short burst, Microsoft'...90.1**0 Posted at: 7 hours ago -
#7933 The maximum decoding length must be > 0:Traceback (most recent call last): File...
💬 1[Last reply:8 hours ago] 知识库中未收录该报错的确切成因，以下为基于报错堆栈和软件机制的可尝试方案：该错误发生在 faster-whisper 的 restore_speech_timestamps（按字级时间戳重新断句）阶段，说明某个语音片段在解码时被计算出的最大解码长度 ≤ 0（通常是片段过短、切片异常或 VAD 切分产生的无效片段）。按顺序尝试：绕过重新断句阶段（最可能有效）在软件根目录（sp.exe 所在文件夹）新...59.36**9 Posted at: 8 hours ago -
#7932 Error during TTS dubbing stage [Edge-TTS (Free)] Dubbing failed completely:403, message='Invalid response status', url='wss://speech.platform.bing.com/consumer/speech/synthesize/re...
💬 1[Last reply:10 hours ago] This is an Edge-TTS rate limit / IP block issue. Edge-TTS (Free) uses Microsoft's public WebSocket endpoint; when too many requests arrive in a short time (or your public IP is flagged), Microsoft's f...37.132**6 Posted at: 10 hours ago -
#7931 翻译字幕阶段出错 [DeepLx] Connection refused, please ensure the local service is started and runningTraceback (most recent call last):...
💬 3[Last reply:11 hours ago] 信息不足，无法判断具体原因。请补充以下内容：软件版本号（主界面左上角，如 v4.15）操作系统（Windows / macOS / Linux，是否源码部署或预打包版）复现步骤：在哪个功能、哪个阶段（语音识别 / 翻译字幕 / 配音 / 合并）报错完整报错信息：软件根目录 logs/ 文件夹下最新的日志文件，复制底部约 30 行报错内容；界面上的红字提示或截图也可补充后可继续为你定位。...106.228**6 Posted at: 11 hours 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.