Footer: show short model slugs from constants

Show resolved model IDs in a subtle footer

Restyle Space card: 🐦 emoji and AIC blue gradient

Fix header logo: embed assets/logo.svg as data URI

Upgrade to Quail Voice Focus 2.2 and aic-sdk 2.5.0

Run dedicated VAD model on voice focus output

Bump aic-sdk to >=2.3.0

Set default value of audio normalization to false

Update default normalizer in compute_wer function to 'english'

Add WER normalization

update-sdk-version (#4)

Specify versions for torchcodec and torch in requirements.txt

default sample value w/ background speech at beginning

vad (#3)

normalization (#2)

dynamic sample rate (#1)

update short_description

update short_description

update short_description

Merge branch 'main' of https://huggingface.co/spaces/ai-coustics/VoiceFocus

display WERs as percentage

Delete .vscode

Update gradio version

Push latest updates

Update offline_pipeline.py

renamed STT names

extended intro and limited file size to 5MB

added download button to all downloadable elements

update soniox streamer to match deepgram

remove redundant description from online.md

ruff

prevent errors when websocket is None

summarize offline enhancement

Add Gradio info message for processing indication to beginning

Update instructions

Refactor online pipeline to improve audio streaming management and enhance clarity in shutdown processes

Refactor offline and online tab structure

Remove debug print statement and add completion message for audio streaming

Update transcribe_audio function to include stream_name parameter and adjust usage in retrieve_audio_information

hide result card earlier

Update documentation for Dawn Chorus and local file upload

Refactor online streaming functionality and enhance documentation

replace 'pytorch' with 'torch' in requirements.txt

add requirements

for debugging in hugginface

typo

remove wer box

rename online to online_pipeline

fix offline pipeline

extra offline file