[BUG] No language specified, language will be first be detected for each audio file (increases inference time).

[mdlaat]

Thank you so much for this project!

Description

Using pluja/whisperx-api:cpu

To Reproduce

Created a working curl command using the /docs endpoint.
The url contains transcription/?model_size=small&language=nl&device=cpu&diarize=true&speaker_min=2&speaker_max=2

Curl command works ok, but in the log of the cotainer it shows:
No language specified, language will be first be detected for each audio file (increases inference time).

Expected behavior

detecting the wanted language (nl)

Environment

• OS client: Ubuntu 22.04
• Browser: Curl
• Version: 7.81.0
• Hosting: Synology DSM 7

Logs and Configuration

No language specified, language will be first be detected for each audio file (increases inference time).