Audio Preference Evaluation

Evaluating preference between two speech modelsLoading...

Mean

Listen to the speech/audio samples and rate which one you prefer.

Which audio would you prefer for Naturalness?

Which audio would you prefer for Intelligibility?

Which audio would you prefer for American Accent?

Which audio would you prefer for Sound Quality?

Comparison Graph

Mean table

Tag	Model 1	Model 2
synthesized	μ: -0.13 / σ: 1.16 / CI95 : 0.05
model1	μ: -0.13 / σ: 1.16 / CI95 : 0.05
model2	μ: -0.13 / σ: 1.16 / CI95 : 0.05

Evaluation

CreatedType

Loading...CUSTOM

Stimulus

CountRequested responsesLanguage

40210 per each audio (10,050 total)English (United States)

Evaluator

Total responsesValid/total evaluatorsGender ratio

10,05010 / 18 (56%)

MaleFemaleOther

82-

Custom evaluation query (5)

Question 1Listen to the speech/audio samples and rate which one you prefer.

Responses

2. A is strongly preferred1. A is preferred0. About the same-1. B is preferred-2. B is strongly preferred

Question 2Which audio would you prefer for Naturalness?

Responses

2. A is strongly preferred1. A is preferred0. About the same-1. B is preferred-2. B is strongly preferred

Question 3Which audio would you prefer for Intelligibility?

Responses

2. A is strongly preferred1. A is preferred0. About the same-1. B is preferred-2. B is strongly preferred

Question 4Which audio would you prefer for American Accent?

Responses

2. A is strongly preferred1. A is preferred0. About the same-1. B is preferred-2. B is strongly preferred

Question 5Which audio would you prefer for Sound Quality?

Responses

2. A is strongly preferred1. A is preferred0. About the same-1. B is preferred-2. B is strongly preferred