我下载的模型是Systran/faster-whisper-large-v3
BTW :V3在huggingface上托管者是systran,而前面的都是Guillaume Klein
然后我看了下这个大佬的github,是苹果法国巴黎工程师,同时是systran的成员,主要贡献是开发了CTranslate2 ,一个用于加速transformers模型推理的组件
以下为模型性能信息
Large-v2 model on GPU
Implementation | Precision | Beam size | Time | Max. GPU memory | Max. CPU memory |
---|---|---|---|---|---|
openai/whisper | fp16 | 5 | 4m30s | 11325MB | 9439MB |
faster-whisper | fp16 | 5 | 54s | 4755MB | 3244MB |
faster-whisper | int8 | 5 | 59s | 3091MB | 3117MB |
Executed with CUDA 11.7.1 on a NVIDIA Tesla V100S.
参考资料
【1】SYSTRAN/faster-whisper: Faster Whisper transcription with CTranslate2 (github.com)
【2】开源语音识别faster-whisper部署教程_faster-whisper-large-v2-CSDN博客
【3】持续进化,快速转录,Faster-Whisper对视频进行双语字幕转录实践(Python3.10) - 刘悦的技术博客 - 博客园 (cnblogs.com)
标签:CTranslate2,faster,部署,whisper,large,GPU,Faster From: https://www.cnblogs.com/YeewahChan/p/18337605