Silxxor/Lucy-addressee-detector

Addressee Detection Model

Fine-tuned BERT for detecting whether speech is directed at the system or not in voice conversations.

Model Description

This model classifies whether a spoken utterance is directed at the voice assistant (addressee) or is ambient speech (talking to yourself, someone else in the room, or just mumbling). Designed for real-time voice chat applications to prevent false triggers.

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

tokenizer = AutoTokenizer.from_pretrained("Silxxor/Lucy-addressee-detector")
model = AutoModelForSequenceClassification.from_pretrained("Silxxor/Lucy-addressee-detector")

text = "tell me the weather for tomorrow"
inputs = tokenizer(text, return_tensors="pt", max_length=64, truncation=True, padding=True)

with torch.no_grad():
    outputs = model(**inputs)
    prediction = torch.argmax(outputs.logits, dim=-1).item()

# 0 = not addressed to system, 1 = addressed to system
print("Addressed to system" if prediction == 1 else "Not addressed to system")

Intended Use

Voice assistants and conversational AI systems that need to distinguish between speech directed at them versus ambient conversation.