WhisperX Slack bot for transcribing audio files
- Accept pyannote/segmentation-3.0 user conditions
- Accept pyannote/speaker-diarization-3.1 user conditions
- Create
HUGGINGFACE_ACCESS_TOKEN
at hf.co/settings/tokens.
- Create an account on Modal
- Save the Hugging Face token on the secrets page, naming it
my-huggingface-secret
, with the keyHUGGINGFACE_ACCESS_TOKEN
. - Get the token id and secret
- Create a new Slack app
- Go to "Event Subscriptions" and subscribe to
message.channels
- Go to "OAuth & Permissions" and add the following scopes:
chat:write
files:read
files:write
channels:read
channels:history
groups:history
im:history
- Get
SLACK_BOT_TOKEN
from "Install App" page (starts withxoxb-
) - Get
SLACK_APP_TOKEN
from "Basic Information" page > "App-Level Tokens" (starts withxapp-
)
- Get
SLACK_BOT_CHANNEL
from the channel you want the bot to watch - Add the bot to the channel
docker run \
-e SLACK_BOT_TOKEN='<YOUR_SLACK_BOT_TOKEN>' \
-e SLACK_APP_TOKEN='<SLACK_APP_TOKEN>' \
-e SLACK_BOT_CHANNEL='<SLACK_BOT_CHANNEL_ID>' \
-e MODAL_TOKEN_ID='<MODAL_TOKEN_ID>' \
-e MODAL_TOKEN='<MODAL_TOKEN_SECRET>' \
ghcr.io/drjuchunkoo/transpal-transcriber:latest