MediumNLPnlp-198

ASR Confidence Average

Problem

Return the token-length-weighted average confidence: sum(len(token) * confidence) / sum(len(token)).

Return 0.0 when there are no tokens or all token lengths are 0.

Hint

Longer tokens weigh more because they contribute more characters.

case 1

in: [["ab","c"],[0.5,1]]

expected: 0.6666666666666666

case 2

in: [["a"],[0.9]]

expected: 0.9

case 3

in: [[],[]]

expected: 0

case 4

in: [["","xy"],[0.1,0.5]]

expected: 0.5

case 5

in: [["ab","c"],[0,0]]

expected: 0

Starter codePython

def asr_confidence_average(tokens, confidences):
    # Your code here
    pass

5 test cases · run your code against them in the editor