We are looking for a web designer for a speech recognition website.
Budget: £1,000 – £4,000 GBP
We are looking for a web designer to assist us with the wireframe and then the design and API integration for a new transcription and speech recognition website.
We are developing a website which will enable our private clinic customers to upload audio files for processing by speech recognition. Three development streams will be required:
a designer to create the wireframe.
a website developer to create the website functionality.
an FTP software specialist to create the FTP functionality for transfers between the website and speech recognition providers and the uploading customers (for which we will provide a template) and APIs to link the website to speech recognition sites such as Deepgram.
Developers/designers will liaise to ensure that the appearance and functionality matches the requirement.
Website functionality
Customers register with the website and specify the service they require. There will be two options:
return files from speech recognition without editing.
return files from speech recognition with editing by DScribe.
* different price points will apply to these options.
Customers would log on to the website and go to ‘Upload Audio’.
Customers then browse on their PC for the files they wish to upload.
Uploaded audio files will be in .wav, .dss or DS2 format. DS2 format files may require decrypting by specialist software such as NCH. This process would be automated where a DS2 file is uploaded.
The uploaded audio files then move to a folder from which they will be pushed/pulled via FTP for processing by AI speech recognition.
Processed documents will be returned via FTP. The returned document must have the same reference as the relevant audio file – ie., UYGB38642.doc, for audio file reference UYGB38642.wav. Initially, DScribe will only return documents in Word format.
Documents for editing (for clients who have selected this option) would move to the ‘Edit’ queue for review by DScribe editors.
Customers may select options for return:
receive files back from speech recognition immediately they are processed.
receive files back once they have been edited.
log on to retrieve files after a specified time.
FTP transfer of files
An API will be developed in conjunction with the speech recognition providers to synchronise transfer of files between the DScribe site and the speech recognition providers. The files would be transferred to the speech recognition providers for processing. On return, the files would follow one of two routes selected by uploading customers:
the file to be returned direct to (or collected by) the customer; or
routed to our team of DScribe editors for review before return or collection by the customer.
DScribe can provide the FTP itself or the developers can develop and incorporate it into the design.
Other requirements:
The website API must pair the uploaded audio file with the completed Word document as well as with the uploading customer. This is to ensure that users cannot access documents uploaded or processed for other customers. This functionality is essential for data protection.
The website must incorporate an audit trail of all uploaded and completed files, sorted by date/time/customer. It should include controls to ensure that files can only be returned (or collected by) the uploading customer and not by any other user.
The site should be able to accept DS2 voice files for processing or include functionality to decrypt DS2 audio files by using NCH software (DS2 files are encrypted).
Completed files should be auto-deleted after 30 days.
On initial login, users should be asked to change their password.
Other functionality can be determined in discussion with the developer. For an example of the format of site we are looking at, look at www.speechtext.AI and www.dictate2us.com
File flow:
Mobile application:
DScribe can use its existing mobile app or a new version can be developed linked to the website. Our preference would be for the latter, depending on lead times and cost. The app will allow users to upload and retrieve files, including allowing editing of text within the application.
Budget:
We are uncertain as to the likely cost of the website and associated applications. We would like to tender this with Freelancer providers, giving them the opportunity come forward with their best cost for the project.
Timeline:
Completion by first week of March 2024.
We are developing a website which will enable our private clinic customers to upload audio files for processing by speech recognition. Three development streams will be required:
a designer to create the wireframe.
a website developer to create the website functionality.
an FTP software specialist to create the FTP functionality for transfers between the website and speech recognition providers and the uploading customers (for which we will provide a template) and APIs to link the website to speech recognition sites such as Deepgram.
Developers/designers will liaise to ensure that the appearance and functionality matches the requirement.
Website functionality
Customers register with the website and specify the service they require. There will be two options:
return files from speech recognition without editing.
return files from speech recognition with editing by DScribe.
* different price points will apply to these options.
Customers would log on to the website and go to ‘Upload Audio’.
Customers then browse on their PC for the files they wish to upload.
Uploaded audio files will be in .wav, .dss or DS2 format. DS2 format files may require decrypting by specialist software such as NCH. This process would be automated where a DS2 file is uploaded.
The uploaded audio files then move to a folder from which they will be pushed/pulled via FTP for processing by AI speech recognition.
Processed documents will be returned via FTP. The returned document must have the same reference as the relevant audio file – ie., UYGB38642.doc, for audio file reference UYGB38642.wav. Initially, DScribe will only return documents in Word format.
Documents for editing (for clients who have selected this option) would move to the ‘Edit’ queue for review by DScribe editors.
Customers may select options for return:
receive files back from speech recognition immediately they are processed.
receive files back once they have been edited.
log on to retrieve files after a specified time.
FTP transfer of files
An API will be developed in conjunction with the speech recognition providers to synchronise transfer of files between the DScribe site and the speech recognition providers. The files would be transferred to the speech recognition providers for processing. On return, the files would follow one of two routes selected by uploading customers:
the file to be returned direct to (or collected by) the customer; or
routed to our team of DScribe editors for review before return or collection by the customer.
DScribe can provide the FTP itself or the developers can develop and incorporate it into the design.
Other requirements:
The website API must pair the uploaded audio file with the completed Word document as well as with the uploading customer. This is to ensure that users cannot access documents uploaded or processed for other customers. This functionality is essential for data protection.
The website must incorporate an audit trail of all uploaded and completed files, sorted by date/time/customer. It should include controls to ensure that files can only be returned (or collected by) the uploading customer and not by any other user.
The site should be able to accept DS2 voice files for processing or include functionality to decrypt DS2 audio files by using NCH software (DS2 files are encrypted).
Completed files should be auto-deleted after 30 days.
On initial login, users should be asked to change their password.
Other functionality can be determined in discussion with the developer. For an example of the format of site we are looking at, look at www.speechtext.AI and www.dictate2us.com
File flow:
Mobile application:
DScribe can use its existing mobile app or a new version can be developed linked to the website. Our preference would be for the latter, depending on lead times and cost. The app will allow users to upload and retrieve files, including allowing editing of text within the application.
Budget:
We are uncertain as to the likely cost of the website and associated applications. We would like to tender this with Freelancer providers, giving them the opportunity come forward with their best cost for the project.
Timeline:
Completion by first week of March 2024.