For our maths training software, we need the numbers from 0 to 1,000 as spoken recordings in Chinese (Mandarin).
How the recordings should be spoken:
- The recordings should be done by a female voice.
- There must be no accent.
- The recordings should be motivating, but not overdone. The users will hear the spoken numbers in every task, i.e. every 15 seconds, so an overdone voice will get on their nerves pretty quickly.
Attached, you find some example of our English recordings as references.
- The recordings must be provided in 1,001 separate wav-files,
- The files must be named like the number they contain, i.e. "0.wav", "1.wav", etc.
- The numbers must not be composed out of sound snippets (i.e. you must not construct "943" by merging the recordings for "900", "40" and "3"). Instead, they must be read individually each.
Estimations of work: We guess that recording itself will take about 5 hours. Postproduction (splitting the recording into individual files) will take another 5 hours.