GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
SayedShaun
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
SayedShaun
Model Tree
openai/whisper-medium
Input Modalities
Output Modalities
Supported Functionality
Stage 1 — ASR:
python
from transformers import pipelineasr = pipeline("automatic-speech-recognition",model="SayedShaun/bengali-whisper-medium",chunk_length_s=20.1) # >20s: competition audio is long-formresult = asr("clip.wav", generate_kwargs={"language": "bn", "task": "transcribe","num_beams": 4, "max_length": 260})raw_text = result["text"] # no punctuation yet
Apply NFC normalization before comparing or storing output — the model emits precomposed Bengali letters (য় ড় ঢ়) that Bengali corpora write as base+nukta; they render identically but compare unequal (cost 18% WER on one otherwise-perfect sample):
python
import unicodedataraw_text = unicodedata.normalize("NFC", raw_text)
Stage 2 — punctuation:
bash
pip install git+https://github.com/sayedshaun/asr-punct-restore.git
python
from asr_punct_restore import PunctuationRestorerrestorer = PunctuationRestorer("SayedShaun/asr-punctuation-restore-bn",layers=(12,)) # most accurate single headpunctuated = restorer(raw_text)
revision so a later push here can't change what you serve.।, ,, ? — no prosody, no exclamation points.Please cite the original author, not this packaging:
bibtex
@misc{tuguldur2023bengaliasr,author = {Tuguldur, Erdene-Ochir},title = {1st place solution, Bengali.AI Speech Recognition},year = {2023},howpublished = {\url{https://www.kaggle.com/competitions/bengaliai-speech/writeups/chimege-1st-place-solution}}}
tugstugi/bengali-ai-asr-submission on Kaggle, mirrored
at bengaliAI/tugstugi_bengaliai-asr_whisper-mediumApache-2.0, following the original upload; the Kaggle submission bundle is CC0.