Cleaned up talk transcript matched to onscreen slides

gemini-talk-transcript.py

from google import genai

import os

client = genai.Client(api_key=os.getenv("GOOGLE_API_KEY"))

# filename_for_upload = "/Users/khkramer/Downloads/maven-lightning-trimmed.mp4"
# myfile = client.files.upload(file=filename_for_upload)
#
# print("My files:")
# for f in client.files.list():
#     print(" ", f.name)
#
# sys.exit()

gfilename = "files/5dd2pvzl3w2o"
prompt = """
This video is a talk about current tools and best practices for building Voice AI agents. Transcribe the talk.
Transcribe the full audio.
Clean up the transcript by removing filler words and adjusting the text to be appropriate to publish in written form. Fix obvious grammatical errors. Improve sentence structure.
For each slide, when the slide first appears insert a short section that describes the slide.
"""

file = client.files.get(name=gfilename)

response = client.models.generate_content(
    # model="gemini-2.0-flash",
    model="gemini-2.5-pro-preview-03-25",
    contents=[
        file,
        prompt,
    ],
)

print(response.text)