Groq TTS (Canopy Labs Orpheus)
Groq retired its PlayAI TTS models in December 2025 and now hosts Canopy Labs' Orpheus models on its OpenAI-compatible speech endpoint. Inputs are capped at 200 characters and output is WAV only.
Overview
Best for: Quick expressive English clips in apps already using Groq; not ideal for long live agent turns.
At a glance
Canopy Labs Orpheus: English ($22/1M) and Saudi Arabic ($40/1M). 6 English voices listed. 200 characters max per request, WAV output only. Vocal direction tags on English only.
Text, max 200 characters, with [directions]
wav only
English; Arabic (Saudi)
6 English, 4 Arabic; no cloning
No numeric claim on Groq's page.
Not specified
Not verified.
Features
- vocal direction tags
- OpenAI-compatible /audio/speech
Pricing
| What | Price | Unit |
|---|---|---|
| orpheus-v1-english | $22 | per 1M characters |
| orpheus-arabic-saudi | $40 | per 1M characters |
900 chars/min; English vs Arabic
Free tier: Groq free tier rate limits apply (not re-verified for TTS)
Source: console.groq.com
Setup
- Create a Groq API key.
- For Arabic, accept the model terms in the Groq console first.
- POST to https://api.groq.com/openai/v1/audio/speech with model, voice, input (<=200 chars).
Endpoint
https://api.groq.com/openai/v1/audio/speechAuthentication
Authorization: Bearer <GROQ_API_KEY>
Quick start python
# pip install groq
import os
from groq import Groq
client = Groq(api_key=os.environ["GROQ_API_KEY"])
resp = client.audio.speech.create(
model="canopylabs/orpheus-v1-english",
voice="troy",
input="[cheerful] Your table is ready. See you soon!", # max 200 chars
response_format="wav",
)
resp.write_to_file("out.wav")Written from the current docs. Check the vendor's SDK version before you ship.
Warnings
200-character limit per request
Every request is capped at 200 characters, so LLM replies must be split into many calls; each adds round-trip latency and counts against rate limits.
PlayAI models are gone
playai-tts and playai-tts-arabic were retired in Dec 2025; old code using them fails.
WAV only, no speed control
Only WAV output is supported and the speed parameter is ignored/unsupported.
No input streaming
Plain request/response over HTTP; no WebSocket TTS.
Plus 12 warnings that apply to all text-to-speech, streaming APIs. See category warnings.
Limits
- 200 characters max per request
- WAV output only
- speed parameter not supported
Models and products
| Name | Status |
|---|---|
| canopylabs/orpheus-v1-english | GA |
| canopylabs/orpheus-arabic-saudi | GA |
| playai-tts / playai-tts-arabic | Deprecated / removed (Dec 2025) |
Docs and sources
Docs
Sources used
- console.groq.com/docs/text-to-speech/orpheus
- console.groq.com/docs/text-to-speech
- releases.sh/release/rel_slohPX8XO2tKT7GyKNH-b-playai-tts-deprecated-orpheus-now...
Exact PlayAI shutdown date (Dec 31 2025 per secondary sources).