Why Is My OpenAI Speech API Input Empty in Python Narration Code?

I'm developing an app that converts descriptive data into narrated audio using OpenAI, but I'm getting an error when creating the audio file. It seems that the input to the OpenAI API is empty, which results in the error message: "String should have at least 1 character"

I suspect the issue is in my parse function, which splits the narration string using "/n" instead of the correct newline character "\n". Below are my code snippets:

Narrater.py

PYTHONCopied
import openai

def parse(narration):
    output = []
    lines = narration.split("/n")
    for line in lines:
        if line.startswith('Background: '):
            background = line.replace("Background:", "Background: ")
            output.append({
                "type":"image",
                "description":background
            })

        elif line.startswith("Narration:"):
            text = line.replace("Narration:", "Narration:")
            output.append({
                "type":"text",
                "content":text
            })
    return output

def create(data,output_path="output_audio.mp3"):
    narration = ""  # Initialize the narration variable as an empty string
    for element in data:
        if element["type"] != 'text':
            continue
        narration += element["content"] + "\n\n"
    audio = openai.audio.speech.create(
        model="tts-1",
        input = narration,
        voice="alloy"
    )
    audio.stream_to_file(output_path)

transcripter.py

PYTHONCopied
# Read the source material from a file
with open("source_material.txt", "r") as f:
    source_material = f.read()

# Create the chat completion
response = openai.chat.completions.create(
    model="gpt-3.5-turbo",
    messages=[
        {
            "role": "system", 
            "content": """You are a short Video Narratorthat reads the data and explains 
            it. You generate 30S TO 1 minutes of narration. 
            The short you create has a background that fades from image to image as the narration is going on.

            Respond in the following manner, repeating until the end of the short:
            Background: Description of background image based on the narration for AI Image Generation and also add the data 
            Narration: A few sentences of narration based on the data in detail  
            
            Background: Description of background image based on the narration for AI Image Generation with data and also add the data 
            Narration: A few sentences of narration based on the data in detail 

            Background: Description of background image based on the narration for AI Image Generation with data and also add the data  
            Narration: A few sentences of narration based on the data in detail

            Background: Description of background image based on the narration for AI Image Generation with data and also add the data 
            Narration: A few sentences of narration based on the data in detail

            Background: Description of background image based on the narration for AI Image Generation with data and also add the data 
            Narration: A few sentences of narration based on the data in detail""",
        },
        {
            "role": "user",
            "content": f"Create a short video narrator based on the following source material:\n\n{source_material}"
        }
    ]
)

# Extract the generated content from the response
generated_content = response.choices[0].message.content

# Print the generated content
data = narrator.parse(generated_content)
print(data)
audio = narrator.create(data, "narration.mp3")

When I run the script, I see that the narration text ends up empty, leading to the error in the audio creation step. This likely means that the newline splitting in the parse function is incorrect.

I have tried downgrading OpenAI to no avail, so I suspect the primary issue is with how I'm processing the generated text. Any guidance on fixing this error would be appreciated!