AI text to speech models: Our pick, Best, and cost
Choose an Audio Studio model before you generate. This page is the chooser. Open Audio Studio when you are ready to run.
Our pick
Qwen Audio 3.0 TTS PlusOur Pick
Qwen Audio 3.0 TTS Plus is Our pick for paid accounts. It is our top speech model for natural reads.
Best
Qwen Audio 3.0 TTS PlusBest
Qwen Audio 3.0 TTS Plus is the Best speech model for us: it sits at the top of the Artificial Analysis Speech Arena for natural, expressive reads.
Most cost-effective
Grok Voice TTS 1.0
Grok Voice TTS 1.0 is the cost pick at 1 credit per 200 characters, with five named voices for everyday voiceovers.
When to choose one over the other
Grok Voice TTS 1.0 vs Gemini 3.1 Flash TTS
Use Grok Voice TTS 1.0 for the lowest credit cost and a small named-voice roster. Use Gemini 3.1 Flash TTS when you want Gemini's larger preset-voice catalog and multilingual scripts.
Every current model
Prices come from the live studio catalog. The selected model stays visible while you work.
Grok Voice TTS 1.0
SpaceXAI speech with five named voices. Default Audio Studio model.
Everyday voiceovers at the lowest credit cost
- Script
- 1 credit / 200 characters
Gemini 3.1 Flash TTS
Google speech with a large preset-voice roster.
Expressive multilingual scripts with Gemini voices
- Script
- 1 credit / 200 characters
Qwen Audio 3.0 TTS Plus
Alibaba's higher-quality Qwen speech model.
Our top speech model for natural reads
- Script
- 1 credit / 200 characters
MiniMax Speech 2.8 HD
Studio-grade MiniMax speech with 17 system voices.
High-fidelity voiceover when quality matters more than cost
- Script
- 2 credits / 200 characters
MAI-Voice-2
Microsoft speech with locale-tagged preset voices.
Scripts that need a specific locale voice
- Script
- 1 credit / 200 characters
Aura-2
Deepgram Aura-2 with a large multilingual preset roster.
Language-specific reads from a wide voice catalog
- Script
- 1 credit / 200 characters
Questions
- Are all voices included?
- Each model ships its own voice roster. Credits are charged per 200 characters, not per voice.
- What does Our pick mean?
- Our pick is the Audio Studio default for paid accounts. Free accounts start on the lowest-cost voice model.
- Where do I generate?
- This page helps you choose. Open Audio Studio to paste a script, pick a voice, and run.
Open Audio Studio
Pick the model and voice, paste the script, and see the credit cost before you generate.