| Frequently asked questions for Speech Technology |
| Q. |
What is speech technology? |
| A. |
Speech technology or voice recognition technology means training computers to recognize the speech of doctors and to have the computers automatically generate the transcript. |
| Q. |
Is it really possible for computers to recognize garbled voices and medical terms of doctors? |
| A. |
Yes, it is possible to train the computers to recognize the medical terms of the doctors; however, while using such technology, even the doctors are trained to dictate as clearly as possible. If the voice is too garbled, then you will need manual intervention. |
| Q. |
How much of work is processed using this technology? |
| A. |
You will be surprised to know that 70% of our workload is already on speech recognition platform, which means we do not use any MTs at all to transcribe 73% of our workload. All this 73% of the volume needs to be proofread by editors. |
| Q. |
What is the average accuracy output of the speech recognition technology? |
| A. |
There will be little-to-no recognition on day #1 when you put a particular doctor’s file thru this technology. However, as time passes, the computer starts recognizing the doctor’s speech pattern and will start producing transcripts ranging from 50-98% accuracy. |
| Q. |
How much time will it take for speech recognition technology for a particular doctor? |
| A. |
It takes anywhere from 0-90 days before a doctor’s speech is recognized and transcripts are automatically produced by the computers, depending on the clarity of the doctor’s speech. |
