Hello all, first I am new to this platform and glad to be part of it, wonderful tool !
My goal is to create an automation that generates instagram video reels scripts based on an audio file which is into a dropbox folder, the result should be into a google doc. The issue is that I only get the transcription of the audio file within the google doc without the “script” part. The “Completion” with the “Create a Prompt Completion GPT” method seems to not work. Here are all the screenshots of my scenario and modules details :
To me, the solution lies in the google doc module setting, it could be the following options : “Choices” or “Usage”. I am looking for further point of views, helps, to solve the issue ! Thank you in advance
Open AI Chat Completions responses take the following form:
{
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"content": "The 2020 World Series was played in Texas at Globe Life Field in Arlington.",
"role": "assistant"
},
"logprobs": null
}
],
"created": 1677664795,
"id": "chatcmpl-7QyqpwdfhqwajicIEznoc6Q47XAyW",
"model": "gpt-3.5-turbo-0613",
"object": "chat.completion",
"usage": {
"completion_tokens": 17,
"prompt_tokens": 57,
"total_tokens": 74
}
}
So the response would be located at the first index of the choices array. (Under message.content)
Hey @DayinNC
ChatGPT module does not actually work similar to how chatgpt interface works
It does not have threads to continue from one module to another
Assistant is an exception
You should map the transcription to create a completion module
In prompt
And then the content from create a completion module will be your desired output
Now you can map that to your google doc
Right now the prompt in create a completion module does not have any content to work with
You will have to give the content of transcription module to this module
Please check within the Choice section. There should be a collection of messages, including the result content. You can map the result in the Google Doc.
Thank you for your detailed answer, here is a screenshot of my Open AI chat completions responses and I don’t know why, the output lenght is only a few words instead of 30 seconds Instagram reels scripts. Here is a screenshot below.
To me, it is due to the Open AI max tokens or settings to modify. Looking for your ideas, thoughts and answers to solve the issue !
Thank you for your detailed answer, here is a screenshot of my Open AI chat completions responses and I don’t know why, the output lenght is only a few words instead of 30 seconds Instagram reels scripts. Here is a screenshot below.
To me, it is due to the Open AI max tokens or settings to modify. Looking for your ideas, thoughts and answers to solve the issue !
Thank you for your detailed answer, here is a screenshot of my Open AI chat completions responses and I don’t know why, the output lenght is only a few words instead of 30 seconds Instagram reels scripts. Here is a screenshot below.
To me, it is due to the Open AI max tokens or settings to modify. Looking for your ideas, thoughts and answers to solve the issue !
In the OpenAI module where you asking it to write the scripts you need to map the text variable from the Whisper output in the module before it to the prompt input in the script writing module.
In your prompt you could do something like this:
summarize the following text in 3 video scripts for Instagram reels of 30 seconds in length each:
