0x16e8bea/protoly-cerebras-voice
0
Protoly Cerebras Voice
Docker Space for testing a realtime voice backend with ASR/TTS hosted on Hugging Face GPU hardware and text generation routed to Cerebras.
Runtime shape:
- Realtime WebSocket:
/v1/realtime - ASR: Parakeet TDT
- TTS: Qwen3-TTS (
Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice, speakerAiden) - LLM: Cerebras Chat Completions via the local Responses API adapter
- Tool calls: forwarded through the adapter and returned as OpenAI Realtime-compatible tool-call events for app-side delegation
Required secret:
CEREBRAS_API_KEY
Optional variables:
CEREBRAS_MODELdefaults togemma-4-31bCEREBRAS_BASE_URLdefaults tohttps://api.cerebras.ai/v1CEREBRAS_RETRY_BUDGET_Sdefaults to14(in-adapter 429 retry window)CEREBRAS_RATE_LIMIT_FALLBACKspoken text when the rate limit persistsQWEN3_TTS_MODELdefaults toQwen/Qwen3-TTS-12Hz-1.7B-CustomVoiceQWEN3_TTS_SPEAKERdefaults toAidenVAD_MIN_SILENCE_MSdefaults to500(lower = snappier turn ends but more speculative LLM requests against the Cerebras quota)
