In addition to ChatGPT's GPT3.5 API update, OpenAI has launched a Whisper speech recognition model. 96 languages are supported.
After installing the openai library in Python, put the audio directory that needs to be translated into it, and run the program to generate the text corresponding to the audio.
import openai
openai.api_key = 'your API Key'
file = open("openai.mp3", "rb")
transcription = openai.Audio.transcribe("whisper-1", file)
translation = openai.Audio.translate("whisper-1", file)
print(transcription, translation)
that's all.