JSON transcript into an SRT file using Machine Learning

I need a software that transforms JSON transcript into an SRT file using Machine Learning I need a freelancer versed in machine learning and statistics to develop a captioning software (it can be a command-line one) which can perform formatting of a force-aligned JSON transcript into an SRT file. It should run on Ubuntu I'm attaching a source file (transcript) and target file (desired outcome) Please let me know how much would you charge for this project. To put you in context, I work a lot on subtitle files and there is a task that is really time consuming for me. Basically what I do is I get a timed transcript (JSON file) of an audio file but I want to transform the JSON transcript into an SRT file with the following format: 1 line per subtitle, and a character length limit of 70 per line. Formatting a subtitles file process should work like this. 1. Select the JSON file 2. Select the function length formatting (x)** 3. The program should split the transcript to create an SRT file. **A line or subtitle must have a length character limit of x (70 characters), x should be editable. I have around 200 input and output files, each of those is around 10 minutes or more, and for each minute I get around 17 captions or subtitles, I'm thinking each subtitle is a unit for the process I want to implement, which is to train the program on splitting sentences at desired words or speaking pauses. So, I'm calculating I currently have around 34,000 "units" for training. Output files might have more words or different words than the input files due to a fixing process, so, ignore them to avoid bugs.