CoolFace
Datasetpublic

Trelis/eval-universal-3-pro-medical-terms-2025-20260408-1928

Evaluation Results: universal-3-pro Evaluation results from Whisper model evaluation. Summary Model WER CER assemblyai/universal-3-pro 6.91% 2.40% Source Data Evaluation Dataset: Trelis/medical-terms-2025 Model Evaluated: assemblyai/universal-3-pro Columns Column Description audio Audio sample (if available from source dataset) reference Ground truth transcription prediction Model prediction wer Word… See the full description on the dataset page: https://huggingface.co/datasets/Trelis/eval-universal-3-pro-medical-terms-2025-20260408-1928.

sourceHugging Faceupdated 6mo agoView on Hugging Face
0likes18downloads
Dataset Card

Evaluation Results: universal-3-pro

Evaluation results from Whisper model evaluation.

Summary

ModelWERCER
assemblyai/universal-3-pro6.91%2.40%

Source Data

Columns

ColumnDescription
audioAudio sample (if available from source dataset)
referenceGround truth transcription
predictionModel prediction
werWord Error Rate for this sample
cerCharacter Error Rate for this sample

| entities | Entity annotations from source dataset | | entity_cer | Per-sample entity CER (-1.0 if no entities) |

Entity CER

Overall Entity CER: 18.06%

CategoryCER
biomarker4.17%
condition3.28%
drug20.33%
procedure12.90%

Generated by [Trelis Studio](https://studio.trelis.com)