Speech Data Evaluation for a Synthesized Voice Engine: Open positions for native English Speakers

Order Details

Description

[Overview]
This is a request for native English speakers to evaluate an English speech data.
Our end goal is to collect a comprehensive evaluation data to improve the quality of a voice(speech) synthesizer.

There are 3 different parts to this evaluation task, and the details are as follows.

[Due date]
2018/6/15(Fri) 18:00(JST)

[Requesting tasks]
　<Basic Information>
　　- Each evaluator will be provided with a "evaluation form" and an individual set of "sound files(.wav)".
　　- The evaluation form contains 50 short phrases as subject of evaluation.
　　- Each set of sound file contains 150 sound files, 3 variations of sound file(VoiceA, VoiceB, VoiceC) for each phrase to be evaluated.

With the materials mentioned above, we request the following 3 tasks.

　1. Comparison and evaluation of 3 voice data variations
　　#Number of positions: 5 unique users　

　2. Cross comparison 1: VoiceA and VoiceB
　　#Number of positions: 10 unique users

　3. Cross comparison 2: VoiceA and VoiceC
　　#Number of positions: 10 unique users

[Remuneration]
　1. Comparison and evaluation of 3 voice data variations...2,250JPY/person
　2. Cross comparison 1: VoiceA and VoiceB...1,500JPY/person
　3. Cross comparison 2: VoiceA and VoiceC...1,500JPY/person

　*Applicants are allowed to fill in for 1 position from each of the 3 tasks.
　In other words, 1 applicant may work on 3 positions across the 3 tasks.

Required Translator Languages

English English → Japanese Japanese → English

Deliverables

excel(.xlsx)

Payment Timing

Immediately after verification