HTTP server that fetches minio audio files and transcribes them using faster_whisper
NOTE: once built, this image does not depend on internet connection.
Run the image:
docker run -p 80:80 \
-e MINIO_ADDRESS=my-minio.com:9000 \
-e MINIO_ACCESS_KEY=my-access-key \
-e MINIO_SECRET_KEY=my-secret-key \
-e MINIO_BUCKET_NAME=my-bucket \
15012002/minio-whisper-api
This will expose an http server listening on 0.0.0.0:80. Here's a request example:
curl --location 'http://localhost/?file=<your_objecct_name>' --no-buffer
MINIO_ADDRESS: Your minio location + port.
MINIO_ACCESS_KEY: Your minio access key. can also be your main user's username.
MINIO_SECRET_KEY: Your minio secret key. can also be yout main user's password.
MINIO_BUCKET_NAME: Your minio's bucket where all the audio is located.
MINIO_SECURE: TRUE if using https, otherwise FALSE (default: FALSE)
Clone this repo.
Build the image:
docker build . -t minio-whisper-api --build-arg="WHISPER_MODEL=medium"
Content type
Image
Digest
sha256:8d867d9fc…
Size
1.9 GB
Last updated
about 2 years ago
docker pull 15012002/minio-whisper-api