OpenAI
Use OpenAI's TTS models to generate natural-sounding speech.
On this page
Requirements
- OpenAI account with billing enabled
- OpenAI API key
- VoiceThisText account with provider connections enabled
Get Your API Key
- Log in at platform.openai.com.
- Open API Keys from the left nav.
- Create a new secret key and copy it once; you cannot view it again.
Connect OpenAI
- In VoiceThisText, go to Providers.
- Click Add Provider and choose OpenAI.
- Paste your OpenAI API key and save.
- VoiceThisText fetches available models and voices automatically.
Models
Select the TTS model you want to use. Common options:
- gpt-4o-mini-tts — fast and cost-effective
- gpt-4o-audio (when available) — higher quality with audio-native output
VoiceThisText shows only the models your key can access. If you upgrade access, click Refresh models in the provider details.
Generating Audio
- Select the OpenAI provider when creating an audio generation.
- Choose a voice and optionally adjust speed or format if shown.
- Generate. VoiceThisText calls OpenAI and stores the audio and transcript timing (if enabled).
- Embed the player or use the WordPress plugin to publish.
Troubleshooting
- 401/insufficient_quota — check billing status, regenerate the API key, and update the provider settings.
- Model not listed — your key may not have access; upgrade your plan and refresh models.
- Rate limits — space out requests or use a lighter model like gpt-4o-mini-tts.
- Voice missing — some voices are model-specific; switch models or refresh voices after OpenAI updates.