Requirements for speechtotext

Overview

Requirements for the speechtotext package:

• Folder with datasets and the transcripts.txt to test the models.

• Contents of .env in the directory of the python script.

Content of .env

Whisper API

OPENAI_API_KEY=sk-somemorerandomnumberlettersandmorerandomnessform
OPENAI_ORGANIZATION=org-somerandomnumberandlette

Amazon transcribe

AWS_ACCESS_KEY_ID=access-id
AWS_SECRET_ACCESS_KEY=acces-key
AMAZON_REGION=eu-west-3
AMAZON_BUCKET=bucket-name

• For bucket with name foo following command returns the region.

curl -sI foo.s3.amazonaws.com | awk '/^x-amz-bucket-region:/ { print $2 }'

Google API

This is the file for an service account. Link to google cloud docs.

GOOGLE_APPLICATION_CREDENTIALS="/path/to/keyfile.json"

Deepgram API

DEEPGRAM_API_KEY=somemorerandomnumberlettersandmorerandomnessform

AssemblyAI API

ASSEMBLY_AI_API_KEY=somemorerandomnumberlettersandmorerandomnessform

Azure API

AZURE_SPEECH_KEY=somemorerandomnumberlettersandmorerandomnessform
AZURE_SPEECH_REGION=somemorerandomnumberlettersandmorerandomnessform

Speechmatics API

SPEECHMATICS_API_KEY=somemorerandomnumberlettersandmorerandomnessform