Design and Program Server Based Transcription Workflow System

We are looking to develop a transcription system to distribute audio files from one server to several different at-home workstations. The files format that these will come in is not know yet but they will be something like wav or mp3's . Typically the files are in some folder structure based on the date but this too is not exactly known. But the files will need to be uploaded in some way such as a daily audio dump onto our server. The data then needs to be parsed so that they are ready to be looked at by various home transcribers. The transcribers should be sent an email letting them know that they have transcription files ready to be worked on. The transcriber then completes their files and these completed files are stored on the server, then the files are ready for a second transcriber to work on...this person will get the same audio files as the first person but also they will get the "first pass" txt files to see and edit as necessary. We want to be able to track any changes that are made between first and second passes through "the tool. " (which may need to be modified for this project) and also we want to have matrixes to look at as far as performance such as speed and accuracy...we should also have an alert to some administrator if a set of transcriptions that are ready for work have not been started within a given time frame. So after the second pass then we can either have a third pass ,or submit to the client depending on how confident we are that the files are as close to 100 percent accurate as possible. So assuming that they are perfect, the system would load back to the client a daily dump of the completed files. We would like for no data to be actually downloaded - audio is streamed from server, and transcribers enter the text while listening to audio tracks and the txt files are also written on server. In addition, there is also the "tool" that they will use. (Currently the tool is written in TCL) We need to develop and integrate an audio transcription tool into this system as well. (I have examples of what it would look like) ## Deliverables We are looking to develop a system to distribute audio files from one server to several different at-home workstations. The files format that these will come in is not know yet but they will be something like wav or mp3's . Typically the files are in some folder structure based on the date but this too is not exactly known. But the files will need to be uploaded in some way such as a daily audio dump onto our server. The data then needs to be parsed so that they are ready to be looked at by various home transcribers. The transcribers should be sent an email letting them know that they have transcription files ready to be worked on. The transcriber then completes their files and these completed files are stored on the server, then the files are ready for a second transcriber to work on...this person will get the same audio files as the first person but also they will get the "first pass" txt files to see and edit as necessary. We want to be able to track any changes that are made between first and second passes through "the tool. " (which may need to be modified for this project) and also we want to have matrixes to look at as far as performance such as speed and accuracy...we should also have an alert to some administrator if a set of transcriptions that are ready for work have not been started within a given time frame. So after the second pass then we can either have a third pass ,or submit to the client depending on how confident we are that the files are as close to 100 percent accurate as possible. So assuming that they are perfect, the system would load back to the client a daily dump of the completed files. We would like for no data to be actually downloaded - audio is streamed from server, and transcribers enter the text while listening to audio tracks and the txt files are also written on server. In addition, there is also the "tool" that they will use. (Currently the tool is written in TCL) We need to develop and integrate an audio transcription tool into this system as well. (I have examples of what it would look like)