Word Error Rate: (S + D + I) / N from Alignment Counts
This lesson is part of the course
In this lesson
You turn alignment counts for eight test utterances into errors, WER and hypothesis length, pool them into a corpus WER and count the utterances above a quality threshold.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- Word Error Rate: (S + D + I) / N from Alignment Counts Locked
- Word-by-Word Comparison: Labelling C, S, D and I Locked
- Edit Distance by Dynamic Programming: the Levenshtein Table Locked
- Character Error Rate: Splitting Strings with MID Locked
- WER by Accent and Noise: SUMIFS, Pooled vs Average Locked
- Real-Time Factor & Latency Percentiles Locked
- Custom Vocabulary: Measuring the Before/After Impact Locked
- Model Comparison: Cost per Audio Hour vs Accuracy Locked