Oriserve/ASR_arena

Update requirements.txt

Update app.py

Added retry logic in fetch audio

Integrated sarvam speech-to-text api

Added ability to upload model and select model option

changed task creation to server

changed model transcription endpoint

changed message to be shown when no text recognized

minor change in instructions

updated abouts page

updated instructions

Changes to about page

Added model descriptions in about and scoreboard page

UI improvements

fixed grammatical errors

changed nan output to empty string

removed duplicate key in scoreboard

removed duplicate key in scoreboard

moved instructions to bottom to better show toast

moved writing to queue to not wait for results to be written

added toast for audio recorded status

Improved status messages using toasts

fixed bug with submit button

gave unique keys to all buttons

converted app to multipage to improve performance

fixed bug for updating dashboard and added preliminary logs

moved writing results to csv to tasks api

fixed issue to allow multiple transcription on one file

fixed requirements.txt

updated ui and file saving logic

bugfixes and added auth

first commit

initial commit