Spaces:
				
			
			
	
			
			
		Build error
		
	
	
	
			
			
	
	
	
	
		
		
		Build error
		
	Add requirements-whisper
Browse files- README.md +5 -4
- dockerfile +1 -1
- requirements-whisper.txt +9 -0
    	
        README.md
    CHANGED
    
    | @@ -52,12 +52,13 @@ You can upload multiple files either through the "Upload files" option, or as a | |
| 52 | 
             
            Each audio file will then be processed in turn, and the resulting SRT/VTT/Transcript will be made available in the "Download" section. 
         | 
| 53 | 
             
            When more than one file is processed, the UI will also generate a "All_Output" zip file containing all the text output files.
         | 
| 54 |  | 
| 55 | 
            -
            ##  | 
| 56 |  | 
| 57 | 
            -
            You can  | 
| 58 | 
            -
            and 2x reduction in memory usage.
         | 
| 59 |  | 
| 60 | 
            -
             | 
|  | |
| 61 | 
             
            ```
         | 
| 62 | 
             
            pip install -r requirements-fastWhisper.txt
         | 
| 63 | 
             
            ```
         | 
|  | |
| 52 | 
             
            Each audio file will then be processed in turn, and the resulting SRT/VTT/Transcript will be made available in the "Download" section. 
         | 
| 53 | 
             
            When more than one file is processed, the UI will also generate a "All_Output" zip file containing all the text output files.
         | 
| 54 |  | 
| 55 | 
            +
            ## Whisper Implementation
         | 
| 56 |  | 
| 57 | 
            +
            You can choose between using `whisper` or `faster-whisper`. [Faster Whisper](https://github.com/guillaumekln/faster-whisper) as a drop-in replacement for the 
         | 
| 58 | 
            +
            default Whisper which achieves up to a 4x speedup and 2x reduction in memory usage. 
         | 
| 59 |  | 
| 60 | 
            +
            You can install the requirements for a specific Whisper implementation in `requirements-fastWhisper.txt` 
         | 
| 61 | 
            +
            or `requirements-whisper.txt`:
         | 
| 62 | 
             
            ```
         | 
| 63 | 
             
            pip install -r requirements-fastWhisper.txt
         | 
| 64 | 
             
            ```
         | 
    	
        dockerfile
    CHANGED
    
    | @@ -14,7 +14,7 @@ RUN apt-get -y install python3-tk | |
| 14 | 
             
            RUN  python3 -m pip install --upgrade pip
         | 
| 15 |  | 
| 16 | 
             
            RUN if [ "${WHISPER_IMPLEMENTATION}" = "whisper" ]; then \
         | 
| 17 | 
            -
                python3 -m pip install -r /opt/whisper-webui/requirements.txt; \
         | 
| 18 | 
             
              else \
         | 
| 19 | 
             
                python3 -m pip install -r /opt/whisper-webui/requirements-fasterWhisper.txt; \
         | 
| 20 | 
             
              fi
         | 
|  | |
| 14 | 
             
            RUN  python3 -m pip install --upgrade pip
         | 
| 15 |  | 
| 16 | 
             
            RUN if [ "${WHISPER_IMPLEMENTATION}" = "whisper" ]; then \
         | 
| 17 | 
            +
                python3 -m pip install -r /opt/whisper-webui/requirements-whisper.txt; \
         | 
| 18 | 
             
              else \
         | 
| 19 | 
             
                python3 -m pip install -r /opt/whisper-webui/requirements-fasterWhisper.txt; \
         | 
| 20 | 
             
              fi
         | 
    	
        requirements-whisper.txt
    ADDED
    
    | @@ -0,0 +1,9 @@ | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | 
|  | |
| 1 | 
            +
            git+https://github.com/huggingface/transformers
         | 
| 2 | 
            +
            git+https://github.com/openai/whisper.git
         | 
| 3 | 
            +
            transformers
         | 
| 4 | 
            +
            ffmpeg-python==0.2.0
         | 
| 5 | 
            +
            gradio==3.23.0
         | 
| 6 | 
            +
            yt-dlp
         | 
| 7 | 
            +
            torchaudio
         | 
| 8 | 
            +
            altair
         | 
| 9 | 
            +
            json5
         |