Our speech recognition system consists of two components:
Scanners. These are instructions for our servers to constantly check email inbox (iMap) or FTP site (via FTP or FTP SSL) or remote directory (via SCH or SFTP) for new audio files. You also setup a delivery instructions here - the transcribed text could be sent via email or SMS (text).
Processors. Our engines will transcribe audio file(s) as soon as they are detected by scanners. When transcription is completed, the resulting text will be handled back to scanners, for the final delivery of the text according to your instructions.
Please note that our processors are capable of generating the confidence score for the transcribed text. And you have an option not to deliver the transcribed text if the confidence level is too low. Our system can process any number of transcription sessions simultaneously, regardless of the audio length.
I have found your service reputable and efficient.