{"id":10359,"date":"2025-05-27T10:01:25","date_gmt":"2025-05-27T08:01:25","guid":{"rendered":"https:\/\/via-internet.de\/blog\/?p=10359"},"modified":"2025-05-27T10:24:38","modified_gmt":"2025-05-27T08:24:38","slug":"daily-how-to-transcribe-videos","status":"publish","type":"post","link":"https:\/\/via-internet.de\/blog\/2025\/05\/27\/daily-how-to-transcribe-videos\/","title":{"rendered":"Daily: How to transcribe Videos"},"content":{"rendered":"\n
\"\"<\/figure>\n\n\n\n

Using Commandline and Python<\/h2>\n\n\n\n

Option 1: Transcribe with Whisper (official)<\/h3>\n\n\n\n
pip install git+https:\/\/github.com\/openai\/whisper.git<\/pre>\n\n\n\n
Then run:<\/p>\n\n\n\n
whisper demo.mp4 --model medium --language auto --output_format txt<\/pre>\n\n\n\n
Optional flags:<\/p>\n\n\n\n
    \n
  • --output_format srt<\/code> (for subtitles)<\/li>\n\n\n\n
  • --language en<\/code> (to skip language detection)<\/li>\n<\/ul>\n\n\n\n
    Create a shell script to transcribe multiple files<\/h4>\n\n\n\n
    #!\/usr\/bin\/env bash\n\n# Loop through all arguments (filenames)\nfor VIDEO in \"$@\"; do\n    echo \"Transcribing: $VIDEO\"\n    whisper \"$VIDEO\" --model medium --language en --output_format txt\ndone<\/pre>\n\n\n\n
    Option 2: Transcribe with faster-whisper (much faster on CPU or GPU)<\/h3>\n\n\n\n
    pip install faster-whisper<\/pre>\n\n\n\n
    Run via Python:<\/p>\n\n\n\n
    import sys\nfrom faster_whisper import WhisperModel\n\nVIDEO=sys.arg[1]\n\nmodel = WhisperModel(\"medium\", device=\"cpu\")  # or \"cuda\" for GPU\nsegments, info = model.transcribe(VIDEO)\n\nwith open(\"transcript.txt\", \"w\") as f:\n    for s in segments:\n        f.write(f\"{s.start:.2f} --> {s.end:.2f}: {s.text.strip()}\\n\")<\/pre>\n\n\n\n
    Optional: Convert to audio (if needed)<\/h3>\n\n\n\n
    If transcription fails or is slow, extract audio first:<\/p>\n\n\n\n
    ffmpeg -i your_video.mp4 -ar 16000 -ac 1 -c:a pcm_s16le audio.wav<\/pre>\n\n\n\n
    Then transcribe audio.wav<\/code> instead.<\/p>\n\n\n\n
    \n\n\n\n
     Problems with numpy<\/h3>\n\n\n\n
    Error when running whisper: A module that was compiled using NumPy 1.x cannot be run in NumPy 2.2.6 as it may crash.<\/p>\n\n\n\n
    Reason:<\/h4>\n\n\n\n
    The error you\u2019re seeing comes from an incompatibility between NumPy 2.x<\/strong> and some Whisper dependencies that were compiled against NumPy 1.x<\/strong>.<\/p>\n\n\n\n
    Solution: Downgrade NumPy<\/h4>\n\n\n\n
    You can fix this by downgrading NumPy to version 1.x<\/strong>:<\/p>\n\n\n\n
    pip install \"numpy<2\"<\/pre>\n\n\n\n
    Then run Whisper again:<\/p>\n\n\n\n
    whisper demo.mp4 --model medium --language auto --output_format txt<\/pre>\n\n\n\n
    Hide warning UserWarning: FP16 is not supported on CPU; using FP32 instead<\/h3>\n\n\n\n
    Option 1: Suppress all Python warnings (quick + global)<\/h3>\n\n\n\n
    In your terminal or script, set the environment variable:<\/p>\n\n\n\n
    PYTHONWARNINGS=\"ignore\" whisper dmeo.mp4 --model medium<br><\/code><\/pre>\n\n\n\n
    Or, in Python code:<\/p>\n\n\n\n
    import warnings\n\nwarnings.filterwarnings(\"ignore\")<\/pre>\n\n\n\n
    Option 2: Suppress only that specific warning<\/h3>\n\n\n\n
    If you’re using faster-whisper<\/code> in Python and want to filter only that one:<\/p>\n\n\n\n
    import warnings\n\nwarnings.filterwarnings(\n    \"ignore\",\n    message=\"FP16 is not supported on CPU; using FP32 instead\"\n)<\/pre>\n\n\n\n
    Option 3: Patch the library (if you’re comfortable)<\/h3>\n\n\n\n
    You can find the line in the faster_whisper<\/code> source code (usually in transcribe.py<\/code>) that issues the warning and comment it out or remove it:<\/p>\n\n\n\n
    # warnings.warn(\"FP16 is not supported on CPU; using FP32 instead\")<\/pre>\n\n\n\n
    Not recommended unless you maintain the code.<\/p>\n","protected":false},"excerpt":{"rendered":"
    Using Commandline and Python Option 1: Transcribe with Whisper (official) Then run: Optional flags: Create a shell script to transcribe multiple files Option 2: Transcribe with faster-whisper (much faster on CPU or GPU) Run via Python: Optional: Convert to audio (if needed) If transcription fails or is slow, extract audio first: Then transcribe audio.wav instead. Problems with numpy Error when running whisper: A module that was compiled using NumPy 1.x cannot be run in NumPy 2.2.6 as it may crash. Reason: The error you\u2019re seeing comes from an incompatibility between NumPy 2.x and some Whisper dependencies that were compiled against NumPy […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_crdt_document":"","_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[1],"tags":[133,171],"class_list":["post-10359","post","type-post","status-publish","format-standard","hentry","category-allgemein","tag-daily","tag-transcribe-video"],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/posts\/10359","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/comments?post=10359"}],"version-history":[{"count":5,"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/posts\/10359\/revisions"}],"predecessor-version":[{"id":10372,"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/posts\/10359\/revisions\/10372"}],"wp:attachment":[{"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/media?parent=10359"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/categories?post=10359"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/via-internet.de\/blog\/wp-json\/wp\/v2\/tags?post=10359"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}