Russian Turn Detection Model (Qwen/Qwen3-0.3B Fine-Tuned)

This model is a specialized End-of-Utterance (EOU) / Turn Detection model designed for Russian spoken dialogue systems. It is fine-tuned from Qwen/Qwen/Qwen/Qwen3-0.6B to classify whether a user has finished speaking or is pausing mid-sentence.

It is optimized for real-time voice agents (like those using LiveKit) to minimize interruptions and reduce latency in conversational flows.

๐ŸŽฏ Model Capabilities

  • Task: Classifies text input as either COMPLETE (user finished) or CONTINUE (user is thinking/pausing).
  • Language: Russian (primary), handles mixed English/Russian technical terms.
  • Latency: Extremely fast inference (based on 0.5B parameter model), suitable for edge or cloud deployment.
  • Nuance: Correctly handles hesitation markers (e.g., "ะฝัƒ...", "ััั...", "ะบะฐะบ ะฑั‹") as CONTINUE.

๐Ÿ’ป Usage

Inference with Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_name = "RAS1981/qwen3-turn-detector-merged"
device = "cuda" if torch.cuda.is_available() else "cpu"

model = AutoModelForCausalLM.from_pretrained(model_name).to(device)
tokenizer = AutoTokenizer.from_pretrained(model_name)

def predict_turn(text):
    messages = [
        {"role": "system", "content": "ะขั‹ ะณะพะปะพัะพะฒะพะน ะฐััะธัั‚ะตะฝั‚. ะžะฟั€ะตะดะตะปัะน, ะทะฐะบะพะฝั‡ะธะป ะปะธ ะฟะพะปัŒะทะพะฒะฐั‚ะตะปัŒ ะณะพะฒะพั€ะธั‚ัŒ."},
        {"role": "user", "content": text}
    ]
    
    inputs = tokenizer.apply_chat_template(
        messages,
        tokenize=True,
        add_generation_prompt=True,
        return_tensors="pt"
    ).to(device)

    outputs = model.generate(
        inputs, 
        max_new_tokens=2, 
        use_cache=True, 
        pad_token_id=tokenizer.eos_token_id
    )
    
    decoded = tokenizer.decode(outputs[0][inputs.shape[1]:], skip_special_tokens=True)
    return decoded.strip()

# Test Cases
print(predict_turn("ะŸั€ะธะฒะตั‚, ั ั…ะพั‡ัƒ ะทะฐะบะฐะทะฐั‚ัŒ ะฟะธั†ั†ัƒ"))  # Output: COMPLETE
print(predict_turn("ะัƒ ั ะดัƒะผะฐัŽ ั‡ั‚ะพ ะผะพะถะตั‚ ะฑั‹ั‚ัŒ..."))   # Output: CONTINUE

๐Ÿ“Š Training Details

Dataset

  • Source: Custom dataset generated via Gemini 2.5 Flash Lite based on IlyaGusev/ru_turbo_alpaca and ss-corpus-ru.
  • Preprocessing:
    • Converted formal text to spoken Russian (added hesitations, fillers, self-corrections).
    • Normalized using NFKC and lowercased specific punctuation.
    • Balanced 50/50 split between COMPLETE and CONTINUE labels to prevent bias.
  • Size: ~400 high-quality curated examples (incremental training).

Hyperparameters

  • Framework: Unsloth + TRL (SFTTrainer)
  • Quantization: 4-bit (QLoRA)
  • Learning Rate: 2e-4
  • Epochs: 62 (Early stopping based on loss convergence)
  • Final Loss: ~0.086
  • Optimizer: AdamW 8-bit

โš ๏ธ Limitations

  • Context: The model looks at the current utterance context. Extremely long pauses in audio might still need VAD (Voice Activity Detection) support.
  • Domain: Fine-tuned on general conversation and real-estate inquiries; may need adaptation for highly specific medical or legal jargon.

๐Ÿ›  Intended Use

  • LiveKit Agents: Use as a semantic turn detector in the EOU plugin.
  • Customer Support Bots: Prevent the bot from interrupting users while they think.
  • Voice Assistants: Improve natural flow in Russian dialogue.
Downloads last month
19
Safetensors
Model size
0.6B params
Tensor type
BF16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for RAS1981/qwen3-turn-detector-merged

Finetuned
Qwen/Qwen3-0.6B
Finetuned
(1,505)
this model
Quantizations
1 model