KailaPavani
/

whisper-small-te

Automatic Speech Recognition

Eval Results (legacy)

Model card Files Files and versions

Metrics Training metrics Community

Finetuned openai/whisper-small on 99 Telugu training audio samples from ./cv-corpus-26.0-2026-06-12/te.

This model was created from the Mozilla.ai Blueprint: speech-to-text-finetune.

Evaluation results on 93 audio samples of Telugu:

Baseline model (before finetuning) on Telugu

Word Error Rate (Normalized): 211.709
Word Error Rate (Orthographic): 197.708
Character Error Rate (Normalized): 174.02
Character Error Rate (Orthographic): 229.888
Loss: 2.799

Finetuned model (after finetuning) on Telugu

Word Error Rate (Normalized): 78.281
Word Error Rate (Orthographic): 102.292
Character Error Rate (Normalized): 48.509
Character Error Rate (Orthographic): 65.316
Loss: 1.08

Downloads last month: 130

Safetensors

Model size

0.2B params

Tensor type

F32

·

Model tree for KailaPavani/whisper-small-te

Base model

openai/whisper-small

Finetuned

(3627)

this model

Evaluation results

wer on Common Voice (Telugu)
self-reported

78.281