I found that in a audio/text realtime stream some instructions force the responses to come back as text instead of audio. Is there a way to force responses to audio?
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Realtime API Audio Modality output | 5 | 231 | November 9, 2024 | |
| Realtime API message response - Audio + Text | 2 | 305 | October 17, 2024 | |
| Even with “modalities” set to “text” only in Realtime API, Audio is occasionally generated | 3 | 362 | November 29, 2024 | |
| Realtime api never sends audio, only text | 1 | 283 | October 17, 2024 | |
| What is the mechanism behind realtime speech to speech api, are transcript and audio stream pushed in a synchronized manner? | 0 | 110 | October 7, 2024 |