Allows for 10 instances to be running simultaneously (requires ~11GB of VRAM at most for all to run) assuming filenames are different

Update README.md

Update README.md

Update README.md

Converted to blocks format to allow for tabs (new microphone tab)

Added padding to batch_decode to handle differing audio sample sizes

Update README.md

added a batch_size input and allow for multiple outputs, also edited the details screen

Remove “lexicon_file” var, it’s no longer needed

Update app.py

Fix bug on line 207

Update requirements.txt

Update requirements.txt

Updated to use Meta's language model

Add a vocab file for Meta's language model.

Add more to detail section

Fix typos

Fix typo

Fix transcript timing issues

Add labels and description

Update to chunking and half precision

Actually fixed the halving code

Fixed the halving precision code

Halving precision to speed up inference

Fixed an error

Trying to fix file downloading

Test 4

Test 3

Test 3

Test 2

First test

initial commit