How Are Speech to Text Minutes Calculated?
When you purchase speech to text minutes for your organization, you are purchasing units in minutes as you would expect. As far as how usage is calculated when you upload audio files to be converted to text, your minutes are calculated and deducted as follows;
For calculations, we convert seconds to 1/100 of a minute. So for example, 15 seconds would be equal to .25 of a minute. (15/60 = .25 minutes)
Seconds are rounded up to the next 15 second interval. So for example, if you have a file that has a length of 3:05, you account is deducted a total of 3.25 minutes. 5 seconds rounded up to 15 seconds equals .25 minutes.
One thing to keep in mind if you are keeping track of you minutes and they don't seem to add up is that milliseconds are factored into the calculations so if you have a file that has a length of 3:00 minutes but you are deducted 3.25 minutes, the reason for this is that the file is most likely not exactly 3:00 minutes but most likely 3:00:xx where it would get rounded up to the next 15 seconds which makes it 3.25.
Here is a table containing some examples and how the deducted minutes are calculated:
Minutes are calculated as part of the back end process when uploading audio files. When you see the minutes on the upload page prior to uploading, this is an estimation. Once a job has been uploaded, the job lister will display the calculated length.
Displayed Audio Length | Actual Audio Length (hh:mm:ss:ms) | Minutes Deducted |
| 5:34 | 00:05:34:23 | 5.75 |
1:03:56 | 01:03:56:01 | 64 |
23:00 | 00:23:00:04 | 23.25 |