Page 78 BBS.pyVideoTrans.com 配音阶段出错 [clone-voice(本地)] {'code': 6, 'msg': 'CUDA error: no kernel image is available for execution on the device CUDA k...

#3345 配音阶段出错 [clone-voice(本地)] {'code': 6, 'msg': 'CUDA error: no kernel image is available for execution on the device CUDA k......

[Last reply:3 months ago] 要解决 RTX 5060 Ti 显卡运行 clone-voice 的 CUDA 内核错误，核心是进入 clone-voice 项目的 Python 环境，安装一个支持新显卡架构的 PyTorch 版本。以下是具体步骤：定位并进入 clone-voice 环境：找到你安装 clone-voice 的目录。如果 clone-voice 使用 Conda 环境，通常目录下会有 environment.y...

138.2* Posted at: 3 months ago

#3344 为什么合成视频特别慢...

💬 1

[Last reply:3 months ago] 视频合成速度慢通常是由于硬件性能不足或软件配置未优化导致的。主要原因及解决方案：未启用 GPU 加速：视频合成（尤其是配音加速/减速、音画对齐、编码）依赖 FFmpeg，默认使用 CPU 运算，速度很慢。检查：确保已安装 CUDA 12.8 和 cuDNN 9.11+。验证：在 CMD 中输入 nvcc -V 和 nvidia-smi 确认 CUDA 版本和显卡状态。配置：在软件“高级选项”中，确...

2408:8226* Posted at: 3 months ago

#3343 RuntimeError: {'message': 'tts failed', 'Exception': 'Could not load libtorchcodec. Likely causes:\n 1. FFmpeg is not properly installed in your environment. We support\n ...

💬 7

[Last reply:3 months ago] 根据截图，服务器上已成功安装 FFmpeg 6.1.1 的共享库（libavutil.so.59 等文件已找到），但 TorchCodec 在加载时仍然失败。这表明问题可能更深层，涉及库文件版本冲突、依赖缺失或 TorchCodec 与当前 PyTorch/CUDA 环境不兼容。解决方案：检查 PyTorch 与 TorchCodec 版本兼容性：错误信息中已提示 PyTorch 版本 (2.10...

2409:896a* Posted at: 3 months ago

#3342 HTTPSConnectionPool(host='translate.google.com', port=443): Max retries exceeded with url: /m?sl=ja&tl=zh-cn&hl=zh-cn&q=%E3%81%8A%E7%96%B2%E3%82%8C%E6%A7%98%E3%81%A7%E3%81%99%E3%80...

💬 1

[Last reply:3 months ago] 这是由于网络连接问题导致无法访问 Google 翻译服务。错误信息表明软件尝试连接 translate.google.com 时失败，可能因为网络环境限制（如需要代理）或 SSL 协议问题。解决方案：配置网络代理：如果您的网络环境需要代理才能访问 Google 服务，请在软件主界面右上角的“网络代理”设置中，填写可用的代理地址（例如 http://127.0.0.1:10809）。更换翻译渠道：G...

43.206* Posted at: 3 months ago

#3341 RuntimeError: {'message': 'tts failed', 'Exception': 'TorchCodec is required for load_with_torchcodec. Please install torchcodec to use this function.'}...

💬 3

[Last reply:3 months ago] 这是由于 GPT-SoVITS 的 API 服务端缺少 torchcodec 依赖或版本不匹配导致的。GPT-SoVITS 服务端需要此库来处理音频编解码。解决方案：检查 GPT-SoVITS API 服务：确保你已正确部署并启动了 GPT-SoVITS 的 API 服务（通常是 http://127.0.0.1:9880 或 http://127.0.0.1:9881）。此错误通常源于服务端环境...

2409:896a* Posted at: 3 months ago

#3340 TaskCfg(cache_folder='D:/win-pyvideotrans-v3.97/tmp/22152/8e0a8641e1', target_dir='D:/Ken Tamplin Vocal Academy - How To Sing Better Than Anyone Else 2.0/Volume 1/_video_out/01. Vo...

💬 1

[Last reply:3 months ago] 这是由于显卡的CUDA计算能力与faster-whisper模型编译的CUDA内核不匹配导致的。错误 cudaErrorNoKernelImageForDevice: no kernel image is available for execution on the device 表明你的NVIDIA显卡（可能是较新的50系显卡）无法运行当前版本的faster-whisper。解决方案：临时方案（...

223.74* Posted at: 3 months ago

#3337 ValueError: Could not get Gradio config from: http://localhost:8000/...

💬 13

[Last reply:3 months ago] 这是由于 Index-TTS2 的 api_server.py 在导入时遇到了更深层的依赖问题，具体是 fairseq 库或其相关模块（这里指向了“text”）无法正确加载。这通常是由于环境依赖冲突或不完整导致的。解决方案：在 Index-TTS2 的云计算环境中，检查并确保安装了所有必需依赖。建议参考 Index-TTS2 项目的官方 README 或 requirements.txt 文件，使...

2409:896a* Posted at: 3 months ago

#3338 Error no file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory E:/AI/pyvideotrans/models/models--Qwen--Qwen3-TTS-1...

💬 1

[Last reply:3 months ago] 这是由于 Qwen3-TTS 模型文件未正确下载或放置导致的。解决方案：确认模型目录：检查路径 E:/AI/pyvideotrans/models/models--Qwen--Qwen3-TTS-12Hz-1.7B-Base 是否存在。手动下载模型：由于 HuggingFace 在国内访问困难，需要手动下载模型文件。从镜像站或可靠来源下载 Qwen3-TTS-12Hz-1.7B-Base 的模型文...

151.247* Posted at: 3 months ago

#3336 使用云算力部署b站index-tts2后执行配音无反应...

💬 3

[Last reply:3 months ago] 这是由于正式执行配音时，软件向 Index-TTS2 API 发送的请求格式或参数与试听测试时不同，导致服务端处理失败或返回了错误但未被日志捕获。根据您提供的截图，API 服务（http://127.0.0.1:8080）已成功启动并显示 Gradio 界面，证明服务本身是运行的。问题核心在于正式执行与试听测试的请求差异。请按以下步骤排查：检查 API 请求格式：正式执行时，软件可能发送了包含多句...

2409:896a* Posted at: 3 months ago

#3335 ValueError: Could not get Gradio config from: http://127.0.0.1:8000/...

💬 1

[Last reply:3 months ago] 这是由于无法连接到本地部署的 F5-TTS 服务导致的。错误表明 pyVideoTrans 尝试从 http://127.0.0.1:8000/ 获取 Gradio 配置信息但失败了。解决方案：确认 F5-TTS 服务已正确启动：确保您已按照官方文档部署并启动了 F5-TTS 服务。服务启动后，应能在浏览器中通过 http://127.0.0.1:8000 访问其 Web 界面。检查服务地址和端口...

2409:896a* Posted at: 3 months ago

#3334 RuntimeError: Unable to open file 'model.bin' in model 'C:/PyVideoTrans/models/models--Systran--faster-whisper-large-v3'...

💬 1

[Last reply:3 months ago] This error indicates that the faster-whisper model file model.bin is missing from the specified directory. This is typically caused by an incomplete or failed model download.Solution Steps:Check the m...

2804:14c* Posted at: 3 months ago

#3333 RuntimeError: Unable to open file 'model.bin' in model 'C:/Users/House Dev/Downloads/PyVideo/models/models--mobiuslabsgmbh--faster-whisper-large-v3-turbo'...

💬 1

[Last reply:3 months ago] This error occurs because the faster-whisper model file (model.bin) is missing from the specified directory. The model download from HuggingFace likely failed or was incomplete.Solution Steps:Manually...

2804:14c* Posted at: 3 months ago

#3332 cannot import name 'GenerationMixin' from 'transformers.generation' (D:\pyvideotrans\_internal\transformers\generation\init.py)...

💬 3

[Last reply:3 months ago] The repeated error indicates the downloaded package itself might be corrupted or your antivirus software is interfering with the extraction/execution of the _internal folder files.Solution Steps:Disab...

162.211* Posted at: 3 months ago

#3331 cannot import name 'GenerationMixin' from 'transformers.generation' (D:\pyvideotrans2\win-pyvideotrans-v3.97 (1)\_internal\transformers\generation\init.py)...

💬 1

[Last reply:3 months ago] This error is caused by an incompatible version of the transformers library within the pre-packaged Windows version of pyVideoTrans.The error cannot import name 'GenerationMixin' indicates t...

162.211* Posted at: 3 months ago

#3330 cannot import name 'GenerationMixin' from 'transformers.generation' (D:\pyvideotrans\win-pyvideotrans-v3.97\_internal\transformers\generation\init.py)...

💬 1

[Last reply:3 months ago] This error is caused by an incompatible version of the transformers library bundled within the pre-packaged Windows version of pyVideoTrans.Solution:Download the latest full package: The issue has bee...

162.211* Posted at: 3 months ago

#3329 TaskCfg(cache_folder='D:/pif/green/win-videotrans-v3.97-LiXianZiMu-ai/tmp/7516/ce3205b571', target_dir='D:/pif/green/win-videotrans-v3.97-LiXianZiMu-ai/output/recogn', remove_noise...

💬 1

[Last reply:3 months ago] 这是由于软件打包环境中的 pytz 模块版本或依赖不完整导致的。pytz 是 pandas 和 modelscope 等库的依赖项，用于处理时区信息。在预打包的 Windows 版本中，该模块可能缺失了必要的子模块 tzinfo。解决方案：临时方案：切换到其他语音识别渠道。在“语音识别渠道”下拉菜单中，选择其他可用的渠道，例如 faster-whisper 或 openai-whisper。这是最...

2409:8a4c* Posted at: 3 months ago

#3328 opening output D:/pyvideotrans/tmp/6164/1771861344.00365-863a7732b4e.zh-Hans.mp3: No such file or directory opening output file D:/pyvideotrans/tmp/6164/1771861344.00365-863a7732b4...

💬 1

[Last reply:3 months ago] 这是由于软件版本过旧（v3.92）导致的临时文件路径处理问题。解决方案：升级软件：这是最直接的解决方案。请下载最新版本 v3.97 的完整包，替换现有程序文件。下载地址：免费下载pyVideoTrans视频翻译软件清理缓存：在升级后，启动软件，点击主界面右上角的“设置”图标，进入“高级设置”，找到并点击“清理缓存”按钮，清除旧的临时文件。规范路径：确保软件解压路径为纯英文、无空格，例如 D:\py...

27.189* Posted at: 3 months ago

#3327 TaskCfg(cache_folder='D:/win-pyvideotrans-v3.97/tmp/2076/47bbb1cd19', target_dir='E:/DownLoads/DCL/_video_out/Breeding My Hot Mommy – Mira Sangre-mp4', remove_noise=False, is_separ...

💬 1

[Last reply:3 months ago] 这是由于使用 openai-whisper (本地) 模型进行语音识别时，模型推理过程中产生了 NaN（非数值）张量，导致解码失败。这通常与模型文件损坏、显存问题或特定视频/音频内容有关。解决方案：切换语音识别渠道：在“语音识别渠道”中，将 openai-whisper(本地) 暂时更换为 faster-whisper(本地)。faster-whisper 通常更稳定且速度更快。检查并重新下载模型...

240e:44d* Posted at: 3 months ago

#3326 How to keep the original voice in the video?...

💬 1

[Last reply:3 months ago] 要保留视频中的原声，需要在配音设置中禁用配音功能。在软件主界面，点击“配音设置”按钮。在配音设置窗口中，将“配音渠道”选择为“No”（即不配音）。确认设置后，执行翻译任务。软件将仅翻译字幕，而不会生成新的配音音频，最终输出的视频将保留原始音轨。请查阅相关文档：如何使用原视频中音色配音 (此文档主要介绍声音克隆，但其中也涉及了保留原声或使用原声进行克隆的概念)...

2001:ee0* Posted at: 3 months ago

#3325 How do I adjust the pause time for comma-like symbols such as (. , ?)?...

💬 1

[Last reply:3 months ago] 这是由于配音时标点符号的停顿时间设置不当导致的。pyVideoTrans 的配音渠道（如 Edge-TTS）会根据标点符号自动插入停顿，但默认的停顿时长可能过长。解决方案：在软件主界面，点击右上角的 “高级选项”。在“高级选项”窗口中，找到 “配音参数” 区域。调整 “逗号停顿秒数” 和 “句号停顿秒数” 这两个参数。默认值可能较大，你可以尝试将其减小（例如，从默认的 0.5 秒调整为 0.2 秒...

2001:ee0* Posted at: 3 months ago

#3345 配音阶段出错 [clone-voice(本地)] {'code': 6, 'msg': 'CUDA error: no kernel image is available for execution on the device CUDA k......

#3344 为什么合成视频特别慢...

#3343 RuntimeError: {'message': 'tts failed', 'Exception': 'Could not load libtorchcodec. Likely causes:\n 1. FFmpeg is not properly installed in your environment. We support\n ...

#3342 HTTPSConnectionPool(host='translate.google.com', port=443): Max retries exceeded with url: /m?sl=ja&tl=zh-cn&hl=zh-cn&q=%E3%81%8A%E7%96%B2%E3%82%8C%E6%A7%98%E3%81%A7%E3%81%99%E3%80...

#3341 RuntimeError: {'message': 'tts failed', 'Exception': 'TorchCodec is required for load_with_torchcodec. Please install torchcodec to use this function.'}...

#3340 TaskCfg(cache_folder='D:/win-pyvideotrans-v3.97/tmp/22152/8e0a8641e1', target_dir='D:/Ken Tamplin Vocal Academy - How To Sing Better Than Anyone Else 2.0/Volume 1/_video_out/01. Vo...

#3337 ValueError: Could not get Gradio config from: http://localhost:8000/...

#3338 Error no file named pytorch_model.bin, model.safetensors, tf_model.h5, model.ckpt.index or flax_model.msgpack found in directory E:/AI/pyvideotrans/models/models--Qwen--Qwen3-TTS-1...

#3336 使用云算力部署b站index-tts2后执行配音无反应...

#3335 ValueError: Could not get Gradio config from: http://127.0.0.1:8000/...

#3334 RuntimeError: Unable to open file 'model.bin' in model 'C:/PyVideoTrans/models/models--Systran--faster-whisper-large-v3'...

#3333 RuntimeError: Unable to open file 'model.bin' in model 'C:/Users/House Dev/Downloads/PyVideo/models/models--mobiuslabsgmbh--faster-whisper-large-v3-turbo'...

#3332 cannot import name 'GenerationMixin' from 'transformers.generation' (D:\pyvideotrans\_internal\transformers\generation\__init__.py)...

#3331 cannot import name 'GenerationMixin' from 'transformers.generation' (D:\pyvideotrans2\win-pyvideotrans-v3.97 (1)\_internal\transformers\generation\__init__.py)...

#3330 cannot import name 'GenerationMixin' from 'transformers.generation' (D:\pyvideotrans\win-pyvideotrans-v3.97\_internal\transformers\generation\__init__.py)...

#3329 TaskCfg(cache_folder='D:/pif/green/win-videotrans-v3.97-LiXianZiMu-ai/tmp/7516/ce3205b571', target_dir='D:/pif/green/win-videotrans-v3.97-LiXianZiMu-ai/output/recogn', remove_noise...

#3328 opening output D:/pyvideotrans/tmp/6164/1771861344.00365-863a7732b4e.zh-Hans.mp3: No such file or directory opening output file D:/pyvideotrans/tmp/6164/1771861344.00365-863a7732b4...

#3327 TaskCfg(cache_folder='D:/win-pyvideotrans-v3.97/tmp/2076/47bbb1cd19', target_dir='E:/DownLoads/DCL/_video_out/Breeding My Hot Mommy – Mira Sangre-mp4', remove_noise=False, is_separ...

#3326 How to keep the original voice in the video?...

#3325 How do I adjust the pause time for comma-like symbols such as (. , ?)?...

Related resource

#3332 cannot import name 'GenerationMixin' from 'transformers.generation' (D:\pyvideotrans\_internal\transformers\generation\init.py)...

#3331 cannot import name 'GenerationMixin' from 'transformers.generation' (D:\pyvideotrans2\win-pyvideotrans-v3.97 (1)\_internal\transformers\generation\init.py)...

#3330 cannot import name 'GenerationMixin' from 'transformers.generation' (D:\pyvideotrans\win-pyvideotrans-v3.97\_internal\transformers\generation\init.py)...