Gandr
Unlimited text-to-speech and voice cloning with flat pricing by audio stream
About Gandr
Gandr provides text-to-speech and voice cloning: create a usable clone from about ten seconds of reference audio, with no training job required and caching after first use. The service supports 23 languages, connects agents via WebSocket, and handles narration and dubbing via HTTP. Every audio clip carries an inaudible watermark, submitted content is not used for training, and official flat pricing is available by audio stream.
Best for
Best use cases
Connect a voice agent via WebSocket
Generate narration via HTTP
Generate dubbing via HTTP
Create a voice clone from a short reference audio clip
Features
Voice Cloning
Create a usable clone from about ten seconds of reference audio, with no training job required; cached after first use.
Multilingual Speech
Speech generation in 23 languages.
Integration
Agents can connect via WebSocket; narration and dubbing can use HTTP.
Content Protection
Each audio clip carries an inaudible watermark; submitted content is not used for training.
Pros
- Create a voice clone from about ten seconds of reference audio
- Supports 23 languages
- Flat pricing by audio stream
- Each clip carries an inaudible watermark, and submitted content is not used for training
Pricing Plans
Free
- Includes 50,000 tokens to start
- 1 token equals 1 character
Take a Gandr
- 1 million tokens per month
- Resets monthly
Scale / flat rate
- 1 unlimited stream
- Unmetered usage
* Pricing information is for reference only. Please refer to the latest announcement on Gandr's official website.