Model reference · open weights
Raon-Speech is an open-weight audio or speech model from KRAFTON, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
Raon-SpeechChat-9B Raon-SpeechChat-9B is a full-duplex speech language model that enables real-time, simultaneous listen-and-speak conversation in English. Built on top of Raon-Speech-9B, it extends the base model with full-duplex decoding — the model can listen to a user and generate speech responses at the same time, supporting natural turn-taking, backchannels ("uh-huh", "mm-hmm"), and barge-in handling. Key Features - Full-Duplex Conversation: Simultaneous listen-and-speak decoding — the model processes user speech and generates responses in real time, just like a natural conversation. - End-to-End Speech Language Model: Built on Qwen3 (36 layers, 4096 hidden dim), Voxtral-Mini-4B-Realtime-2602 Audio Encoder (32 layers), Mimi codec (32 quantizers), ECAPA-TDNN speaker encoder, Qwen3OmniMoeTalkerCodePredictor (5 layers, 1024 hidden dim), and Qwen3-based Talker (4 layers, 2048 hidden dim). - Backchannel Responses: Dedicated backchannel token (<|audiooutputbackchannel|) for natural conversational feedback like "uh-huh" and "mm-hmm", with adjustable frequency via backchannel penalty. - Speak-First / Listen-First Modes: Configurable via runtime token forcing — the model can either wait for user speech before responding (listen-first) or begin speaking immediately (speak-first). - Persona-Driven Conversations: 17 built-in personas with customizable system prompts, context injection, and persona catalog support. - Speaker Voice Conditioning: Optional speaker reference audio for voice cloning via ECAPA-TDNN embeddings. - HuggingFace Transformers Integration: Load and run directly via AutoModel.frompretrained with trustremotecode=True — no custom package installation required. Benchmark Results Raon-SpeechChat performs strongly on conversational speech capabilities such as pause handling, backchanneling, smooth turn-taking, interruption handling, overlap robustness, and multi-turn dialogue. Requirements Quick Start Option 1: Load from Hub (recommended) No pip install raon needed. Option 2: With raon package installed Self-Hosted Demo Run the full-duplex speech conversation demo locally in your browser with Docker. Prerequisites: NVIDIA GPU with CUDA 12.x (16 GB+ VRAM)
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | KRAFTON |
|---|---|
| Type | Audio & music |
| Parameters (lead) | 9.7B |
| Variants | 1 |
| Runs with | transformers |
| Released | 2026-04-01 |
| Popularity | 1k downloads / month |
| Likes | 46 |
| Licence | Commercial licence needed |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| Raon-SpeechChat-9B | 9.7B | BF16 | ~22.3 GB | ✓ | Weights ↗ |
Using it via the API
Once AxForge deploys raon-speech for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (raon-speech below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \ -H "Authorization: Bearer $AXFORGE_API_KEY" \ -F model="raon-speech" -F file=@audio.mp3
Licence
The weights are open but cc-by-nc-4.0 needs a commercial agreement for business use. AxForge can arrange that licence and host the model for you — you pay AxForge, we settle with the model’s maker. Ask us for a quote. Read the licence ↗