# Hugging Face and Cerebras bring Gemma 4 to real-time voice AI

Source: [huggingface.co](https://huggingface.co/blog/cerebras-gemma4-voice-ai)  
Feed7 permalink: https://feed7.dev/p/cerebras-gemma4-voice-ai-1v86fff  
Published: Unknown  
Trust: Source Linked (source_linked)

## Why Included

Hugging Face demos real-time speech-to-speech from open parts: Nvidia Parakeet ASR, Gemma 4 31B on Cerebras inference, Alibaba's Qwen3TTS — pipeline code is open and already runs on 9,000+ Reachy Mini robots.

## Source Summary

**The gist** Hugging Face and Cerebras show a real-time voice pipeline built entirely from open components: Nvidia's **Parakeet** for speech recognition, **Gemma 4 31B** running on Cerebras inference, and Alibaba's **Qwen3TTS** for output. The code is in the **huggingface/speech-to-speech** repo with a Spaces demo, and the stack already ships on **9,000+ Reachy Mini** robots.

## Practical Implication

**Why it matters** A modular ASR-to-LLM-to-TTS chain on a fast inference provider is now a copyable recipe for adding voice to an agent product without a proprietary end-to-end voice model — each stage swaps independently, so you can trade model quality against latency per stage.

## Agent-Ready Context

**The gist** Hugging Face and Cerebras show a real-time voice pipeline built entirely from open components: Nvidia's **Parakeet** for speech recognition, **Gemma 4 31B** running on Cerebras inference, and Alibaba's **Qwen3TTS** for output. The code is in the **huggingface/speech-to-speech** repo with a Spaces demo, and the stack already ships on **9,000+ Reachy Mini** robots.

**Why it matters** A modular ASR-to-LLM-to-TTS chain on a fast inference provider is now a copyable recipe for adding voice to an agent product without a proprietary end-to-end voice model — each stage swaps independently, so you can trade model quality against latency per stage.

**Watch out** The post claims low latency but publishes **no benchmarks** — no tokens per second, no **P95 latency** figures — and no Cerebras pricing. Measure the pipeline on your own workload before depending on it.

## Context Map

- Layer: model
- Domains: audio
- Topics: open-models

## Uncertainty

- The post claims low latency but publishes **no benchmarks** — no tokens per second, no **P95 latency** figures — and no Cerebras pricing. Measure the pipeline on your own workload before depending on it.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.