Why Whisper cuts off Indic transcripts after six seconds
• 5
AI & ML interests
Speech and NLP systems for Indian legal infrastructure
Recent Activity
Papers
SCRIBE: Diagnostic Evaluation and Rich Transcription Models for Indic ASR
Vividh-ASR: A Complexity-Tiered Benchmark and Optimization Dynamics for Robust Indic Speech Recognition
Articles
kavyamanohar
posted an update about 21 hours ago
Post
30
Some bugs teach you more than they cost you. It's a small bug with a big lesson I keep running into: speech tools are built and benchmarked on English, so their limits are quietly calibrated for English. You only find the edges when you work in the languages they weren't tested on.
Learnt things the hard way, wrote it down so you don't have to.
https://huggingface.co/blog/adalat-ai/whisper-token-limit
Learnt things the hard way, wrote it down so you don't have to.
https://huggingface.co/blog/adalat-ai/whisper-token-limit
karynaur
updated a
collection 1 day ago
kavyamanohar
published an article 4 days ago
Article
Why Whisper cuts off Indic transcripts after six seconds
adalat-ai
• • 5model release?
👀 1
3
#1 opened about 1 month ago
by 
trysem
kavyamanohar
published 3
models 22 days ago
adalat-ai/whisper-small-hi-curated-reverse-mft-1-1-1
Automatic Speech Recognition • 0.2B • Updated • 18
adalat-ai/whisper-small-kn-curated-reverse-mft-1-1-1
Automatic Speech Recognition • 0.2B • Updated • 18
adalat-ai/whisper-small-ml-curated-reverse-mft-1-1-1
Automatic Speech Recognition • 0.2B • Updated • 23
kavyamanohar
updated a
collection 24 days ago
kavyamanohar
updated a
model 24 days ago
kavyamanohar
authored a
paper about 2 months ago
Kush0610
authored a
paper about 2 months ago
orgh0
authored a
paper about 2 months ago
kavyamanohar
authored a
paper about 2 months ago