Canopy Labs’ Orpheus TTS is live on GroqCloud
Today we’re announcing support for Canopy Labs’ Orpheus text-to-speech (TTS) on GroqCloud, with two model variants built for real-time, high-quality voices:
- English TTS: canopylabs/orpheus-v1-english (with vocal directions)
- Saudi Arabic (dialect) TTS: canopylabs/orpheus-arabic-saudi (authentic pronunciation + regional nuance)
These Orpheus models serve as the TTS replacement on GroqCloud for both PlayAI-TTS and PlayAI-TTS-Arabic, helping builders move to a more expressive English experience and higher-quality, more natural Saudi Arabic voices.
Why Orpheus on GroqCloud
Voice apps only feel “human” when they’re fast, natural, and reliable, especially for voice agents, customer support, and interactive experiences. GroqCloud’s TTS endpoint is designed to convert text to audio in seconds, with models tuned for English expressiveness and Saudi Arabic authenticity.
What you get out of the box:
- OpenAI-compatible speech endpoint: https://api.groq.com/openai/v1/audio/speech
- Two specialized Orpheus models (English + Arabic Saudi) hosted on GroqCloud
- Predictable, character-based pricing per 1M characters
- Low-latency inference and multiple voice personas (6 English voices, 4 Saudi Arabic voices)
Current performance: both models are currently delivering up to ~100 characters/second.
Orpheus V1 English: Expressive Speech with “Vocal Directions”
Orpheus V1 English is an expressive TTS model that supports six professionally-trained English voices as well as bracketed vocal directions, so you can steer delivery with tags like [cheerful] or [whisper].
It was trained on 100k+ hours of English speech and billions of text tokens, enabling human-level speech generation while maintaining strong language understanding.
Orpheus Arabic Saudi: Authentic Saudi Dialect Synthesis
Orpheus Arabic Saudi is a specialized text-to-speech model developed by Canopy Labs that generates authentic Saudi dialect speech with natural pronunciation and regional nuances. This model offers four distinct Saudi dialect voices and low-latency inference, optimized for applications requiring high-quality Arabic speech synthesis.
Note: vocal directions are not supported for this model at this time.
Use Cases: From Voice Agents to Creative Production
Orpheus is built for high-quality, low-latency TTS workloads like that of:
- Voice agent experiences: natural conversational speech for interactive apps and dynamic dialogue flows
- Customer support + accessibility: lifelike voices for support systems and assistive tools, in English and Saudi Arabic
- Creative content generation: narration, storytelling, character voices, and content localization
Pricing (per 1M characters)
Character-based pricing keeps costs predictable as you scale:
- canopylabs/orpheus-v1-english : $22 / 1M characters
- canopylabs/orpheus-arabic-saudi : $40 / 1M characters
Build fast
Orpheus on GroqCloud makes it easy to ship real-time voice, from expressive English narration to authentic Saudi Arabic speech, without compromising latency or cost predictability. Get started with canopylabs/orpheus-v1-english and canopylabs/orpheus-arabic-saudi via the GroqCloud Developer Console, available in the Playground and the API and take low-latency TTS workloads to the next level!
To learn more about Orpheus TTS and best practices check out our developer documentation.