File size: 9,149 Bytes
c7e434a e05726c 4480a83 c7e434a e05726c c7e434a 4480a83 c7e434a 4480a83 c7e434a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 |
"""
Structure Analysis Service
Analisis struktur berbicara (opening, content, closing)
"""
import pandas as pd
import torch
import re
from transformers import AutoTokenizer, AutoModelForSequenceClassification
from typing import List, Dict
from app.core.device import get_device
class StructureService:
"""Analisis struktur public speaking"""
def __init__(self, model_path: str = 'Cyberlace/swara-structure-model'):
"""
Initialize model from Hugging Face Hub
Args:
model_path: HF Hub model name or local path
"""
print("📊 Initializing Structure Service...")
print(f"📦 Loading model from: {model_path}")
# Auto-detect device
self.device = get_device()
# Load from Hugging Face Hub (with caching)
self.tokenizer = AutoTokenizer.from_pretrained(
model_path,
cache_dir="/.cache"
)
self.model = AutoModelForSequenceClassification.from_pretrained(
model_path,
cache_dir="/.cache"
)
self.model.to(self.device) # Move model to device
self.model.eval()
self.label_map = {0: 'opening', 1: 'content', 2: 'closing'}
print("✅ Structure Service ready!\n")
def split_into_sentences(self, text: str) -> List[str]:
"""Split text menjadi kalimat-kalimat"""
sentences = re.split(r'[.!?,;\n]+', text)
sentences = [s.strip() for s in sentences if s.strip()]
return sentences
def predict_sentences(self, sentences: List[str], confidence_threshold: float = 0.7) -> List[Dict]:
"""Prediksi label untuk list kalimat"""
results = []
for idx, sentence in enumerate(sentences):
inputs = self.tokenizer(
sentence,
add_special_tokens=True,
max_length=128,
padding='max_length',
truncation=True,
return_tensors='pt'
)
# Move inputs to device
inputs = {k: v.to(self.device) for k, v in inputs.items()}
with torch.no_grad():
outputs = self.model(**inputs)
probs = torch.nn.functional.softmax(outputs.logits, dim=-1)
predicted_class = torch.argmax(probs, dim=-1).item()
confidence = probs[0][predicted_class].item()
predicted_label = self.label_map[predicted_class]
# Jika opening/closing tapi confidence rendah → ubah jadi content
if predicted_label in ['opening', 'closing'] and confidence < confidence_threshold:
predicted_label = 'content'
results.append({
'sentence_idx': idx,
'text': sentence,
'predicted_label': predicted_label,
'confidence': confidence
})
return results
def apply_structure_rules(self, predictions: List[Dict]) -> List[Dict]:
"""Terapkan rules untuk memperbaiki struktur"""
if not predictions:
return predictions
n = len(predictions)
# Define keywords first (will be used in rules)
closing_keywords = [
'demikian', 'terima kasih', 'terimakasih', 'sekian', 'akhir kata',
'wassalamualaikum', 'wassalam', 'waalaikumsalam',
'sampai jumpa', 'sampai bertemu', 'salam penutup'
]
opening_keywords = [
'selamat pagi', 'selamat siang', 'selamat sore', 'selamat malam',
'assalamualaikum', 'assalamu alaikum',
'hadirin yang', 'bapak ibu', 'pertama-tama', 'izinkan saya',
'perkenalkan', 'yang terhormat'
]
# Rule 1: 2 kalimat pertama cenderung opening (HANYA jika ada opening keyword)
for i in range(min(2, n)):
text_lower = predictions[i]['text'].lower()
has_opening_kw = any(kw in text_lower for kw in opening_keywords)
if has_opening_kw and predictions[i]['confidence'] > 0.5:
predictions[i]['predicted_label'] = 'opening'
# Rule 2: 2 kalimat terakhir cenderung closing (HANYA jika ada closing keyword)
for i in range(max(0, n-2), n):
text_lower = predictions[i]['text'].lower()
has_closing_kw = any(kw in text_lower for kw in closing_keywords)
if has_closing_kw and predictions[i]['confidence'] > 0.5:
predictions[i]['predicted_label'] = 'closing'
# Rule 3: Keyword detection untuk semua kalimat (override model prediction)
for pred in predictions:
text_lower = pred['text'].lower()
# Check OPENING first (lebih prioritas untuk kalimat awal)
is_opening_keyword = any(kw in text_lower for kw in opening_keywords)
# Check CLOSING - tapi EXCLUDE jika ada opening keyword
# Ini prevent "assalamualaikum" salah dideteksi sebagai closing karena "salam"
is_closing_keyword = any(kw in text_lower for kw in closing_keywords)
if is_opening_keyword and not is_closing_keyword:
pred['predicted_label'] = 'opening'
elif is_closing_keyword and not is_opening_keyword:
pred['predicted_label'] = 'closing'
return predictions
def segment_speech_structure(self, predictions: List[Dict]) -> Dict:
"""Grouping kalimat berdasarkan struktur"""
structure = {
'opening': [],
'content': [],
'closing': []
}
for pred in predictions:
label = pred['predicted_label']
structure[label].append(pred)
return structure
def calculate_score(self, structure: Dict) -> Dict:
"""Hitung skor berdasarkan struktur"""
has_opening = len(structure['opening']) > 0
has_content = len(structure['content']) > 0
has_closing = len(structure['closing']) > 0
if has_opening and has_content and has_closing:
score = 5
description = "Sempurna! Struktur lengkap (Pembuka, Isi, Penutup)"
elif has_opening and has_content and not has_closing:
score = 4
description = "Baik. Ada pembuka dan isi, tapi kurang penutup"
elif has_opening and not has_content and has_closing:
score = 3
description = "Cukup. Ada pembuka dan penutup, tapi isi kurang jelas"
elif not has_opening and has_content and has_closing:
score = 2
description = "Perlu perbaikan. Kurang pembuka yang jelas"
elif has_opening and not has_content and not has_closing:
score = 1
description = "Kurang lengkap. Hanya ada pembuka"
else:
score = 0
description = "Struktur tidak terdeteksi dengan baik"
return {
'score': score,
'max_score': 5,
'description': description,
'category': description.split('.')[0] if '.' in description else description,
'has_opening': has_opening,
'has_content': has_content,
'has_closing': has_closing,
'opening_count': len(structure['opening']),
'content_count': len(structure['content']),
'closing_count': len(structure['closing'])
}
def analyze(self, transcript: str, apply_rules: bool = True) -> Dict:
"""
Analisis struktur speech
Args:
transcript: Teks lengkap dari speech
apply_rules: Gunakan heuristic rules
Returns:
Dict berisi hasil analisis
"""
print(f"📊 Analyzing structure...")
# Split into sentences
sentences = self.split_into_sentences(transcript)
# Predict
predictions = self.predict_sentences(sentences)
# Apply rules
if apply_rules:
predictions = self.apply_structure_rules(predictions)
# Segment structure
structure = self.segment_speech_structure(predictions)
# Calculate score
score_result = self.calculate_score(structure)
print("✅ Structure analysis complete!\n")
return {
'score': score_result['score'],
'category': score_result['category'],
'description': score_result['description'],
'has_opening': score_result['has_opening'],
'has_content': score_result['has_content'],
'has_closing': score_result['has_closing'],
'opening_count': score_result['opening_count'],
'content_count': score_result['content_count'],
'closing_count': score_result['closing_count'],
'total_sentences': len(sentences)
}
|