Fixing translation mistakes
As you saw, translations aren't always perfect, particularly if the transcript contains names or references outside of the model's knowledge. In this exercise, you'll use another call to the chat model to fix these incorrect words.
The English transcript you created in the previous exercise is available as en_transcript.
This exercise is part of the course
Multi-Modal Systems with the OpenAI API
Exercise instructions
- Create a prompt that fixes the technical names and terminology in the English transcript,
en_transcript, leaving the rest of the transcript unchanged.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
client = OpenAI(api_key="")
# Fix the mistake in the transcript
response = ____
corrected_text = response.choices[0].message.content
print(corrected_text)