refactor: QQ → OBv11 重命名 + 平台格式统一抽象

- 所有对外称呼从 QQ 改为 OBv11（注释/提示词/日志/配置项） - 新增 PlatformFormat 结构体，统一管理平台消息标记格式 - defaultPlatformFormats() 注册表替代硬编码 qqTargetRe - extractProactiveMessage 改为 Thinker 方法，遍历格式注册表匹配 - 配置项重命名: QQ_BOT_PORT → OBV11_BOT_PORT, QQBotPort → OBv11BotPort - 标记格式: 【QQ群聊】→【OBv11群聊】、【QQ私聊】→【OBv11私聊】 Co-Authored-By: Claude <noreply@anthropic.com>
2026-06-22 20:48:07 +08:00
parent 70dbb23234
commit 41f653b672
25 changed files with 1609 additions and 82 deletions
@@ -0,0 +1,88 @@
+#!/usr/bin/env python3
+"""K-means 聚类自动分组声纹，每组抽 1 个样本供试听确认"""
+import os, sys, json, warnings, shutil
+import numpy as np
+import librosa
+from sklearn.cluster import KMeans
+warnings.filterwarnings('ignore')
+
+SEARCH_DIR = r"D:\Project\Code\Uni\Cyrene-Voice-Model\data\cleaned"
+SAMPLE_DIR = r"D:\Project\Code\Uni\Cyrene-Voice-Model\data\voice_samples"
+N_CLUSTERS = 8  # 预期 3-5 个女声 + 若干男声/杂音组
+
+def extract_features(wav_path):
+    try:
+        y, sr = librosa.load(wav_path, sr=22050, mono=True)
+        if len(y) < sr * 0.3: return None
+        f0, _, _ = librosa.pyin(y, fmin=80, fmax=600, sr=sr)
+        f0 = f0[~np.isnan(f0)]
+        if len(f0) < 10: return None
+        mfcc = librosa.feature.mfcc(y=y, sr=sr, n_mfcc=13)
+        mfcc_d = librosa.feature.delta(mfcc)
+        feat = np.concatenate([
+            [np.mean(f0), np.std(f0),
+             np.percentile(f0, 25), np.percentile(f0, 50), np.percentile(f0, 75)],
+            np.mean(mfcc, axis=1), np.std(mfcc, axis=1),
+            np.mean(mfcc_d, axis=1), np.std(mfcc_d, axis=1),
+        ])
+        return feat.astype(np.float32)
+    except:
+        return None
+
+print("提取声纹特征...")
+wav_files = []
+features = []
+for root, dirs, files in os.walk(SEARCH_DIR):
+    for f in files:
+        if f.endswith('.wav') and 'VoBanks' in root:
+            path = os.path.join(root, f)
+            feat = extract_features(path)
+            if feat is not None:
+                wav_files.append(path)
+                features.append(feat)
+
+features = np.array(features)
+print(f"  有效文件: {len(features)}")
+
+# K-means 聚类
+print(f"\nK-means 聚类 (k={N_CLUSTERS})...")
+kmeans = KMeans(n_clusters=N_CLUSTERS, random_state=42, n_init=10)
+labels = kmeans.fit_predict(features)
+
+# 统计每组
+clusters = {}
+for i, (label, path) in enumerate(zip(labels, wav_files)):
+    if label not in clusters:
+        clusters[label] = []
+    clusters[label].append((path, features[i]))
+
+# 每组选最接近中心的样本
+print(f"\n=== 聚类结果 ===\n")
+for label in sorted(clusters.keys()):
+    group = clusters[label]
+    center = kmeans.cluster_centers_[label]
+
+    # 找离中心最近的
+    best_idx = min(range(len(group)), key=lambda i: np.linalg.norm(group[i][1] - center))
+    best_path = group[best_idx][0]
+
+    # 统计音高
+    pitches = [g[1][0] for g in group]  # mean pitch
+    avg_pitch = np.mean(pitches)
+
+    voice_type = "男" if avg_pitch < 170 else "女"
+    print(f"  Group {label+1}: {len(group):4d} files, pitch={avg_pitch:.0f}Hz ({voice_type}), "
+          f"sample: {os.path.basename(best_path)}")
+
+# 复制每组样本到样本目录
+os.makedirs(SAMPLE_DIR, exist_ok=True)
+for label in sorted(clusters.keys()):
+    group = clusters[label]
+    center = kmeans.cluster_centers_[label]
+    best_idx = min(range(len(group)), key=lambda i: np.linalg.norm(group[i][1] - center))
+    src = group[best_idx][0]
+    dst = os.path.join(SAMPLE_DIR, f"group_{label+1:02d}_{os.path.basename(src)}")
+    shutil.copy2(src, dst)
+
+print(f"\n每组样本已复制到: {SAMPLE_DIR}")
+print("试听每个 group_*.wav，找到昔涟的组，告诉我编号。")