When managing large collections of audio files, automating the transcription process can save significant time and effort. This guide explains how to automate transcription of audio files using Deepgram's Nova model on a Windows system. By following these steps, users can efficiently convert .m4a, .aac, and .opus files into transcriptions with diarization.
Before proceeding, ensure you have the following prerequisites:
You can create a batch file that automates the process of uploading audio files to Deepgram and saving the transcriptions.
Here's an example:
@echo off
setlocal enableextensions
set "api_key=YOUR_DEEPGRAM_API_KEY"
set "model=nova"
for %%f in ("*.m4a" "*.aac" "*.opus") do (
echo Processing "%%~nf"
curl -X POST \
--data-binary @"%%f" \
-H "Content-Type: audio/m4a" \
-H "Authorization: Token %api_key%" \
"https://api.deepgram.com/v1/listen?model=%model%" > "%%~nf.txt"
)
endlocal
echo Transcription complete.
pause
Replace YOUR_DEEPGRAM_API_KEY
with your actual Deepgram API key.
Adjust Content-Type Header based on the file type if necessary (e.g., audio/aac
for AAC files).
Save the file with a .bat
extension (e.g., transcribe.bat
).
.txt
format with diarization.for
loop processes files sequentially; large batches may take some time.By following these instructions, users can swiftly automate the transcription of their audio files using Deepgram's API, saving both time and effort. This process is ideal for anyone regularly working with bulk audio data and requiring efficient transcription.