OpenAI's text-to-speech (TTS)
OpenAI now provide models for creating human-like speech from a text input, so-called text-to-speech or TTS. OpenAI provide several voices to choose from, and they provide the ability to stream the response to local files or downstream applications.
This exercise is part of the course
Multi-Modal Systems with the OpenAI API
Exercise instructions
- Create the text-to-speech request for
"Hi! How's your day going?", using the"ballad"voice. - Stream the response to an
.mp3file.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
client = OpenAI(api_key="")
# Create the text-to-speech request
response = client.audio.speech.create(
model="gpt-4o-mini-tts",
____,
input="Hi! How's your day going?"
)
# Stream the response to an MP3 file
response.____("output.mp3")