feat: integrate TTS voice responses and improve response loading UX

.env.example

@@ -1,71 +1,88 @@
+# =============================================================================
+# Alya Bot - Environment Configuration
 # Copy this file to .env and fill in your actual values
-# ============================================================================
+# =============================================================================
+
+# =============================================================================
+# BOT (REQUIRED)
+# =============================================================================
 
 # Get your bot token from: https://t.me/BotFather
-TELEGRAM_BOT_TOKEN=your_telegram_bot_token_here  # Replace with your actual Telegram Bot Token
+TELEGRAM_BOT_TOKEN=your_telegram_bot_token_here
 
-# ============================================================================
-# ADMIN CONFIGURATION (REQUIRED)
-# ============================================================================
 # Your Telegram User ID (get from: https://t.me/userinfobot)
-# For multiple admins, use comma-separated list: 12345678,87654321
-ADMIN_IDS=your_telegram_ids  # Comma-separated list of Telegram user IDs, e.g., 12345678,87654321
+# Multiple admins: comma-separated, e.g. 12345678,87654321
+ADMIN_IDS=your_telegram_user_id
+
+# =============================================================================
+# GEMINI API (REQUIRED)
+# =============================================================================
 
-# ============================================================================
-# GOOGLE GEMINI API CONFIGURATION (REQUIRED)
-# ============================================================================
 # Get API keys from: https://aistudio.google.com/api-keys
-# Multiple keys supported for automatic rotation and reliability
-GEMINI_API_KEYS=gemini_key_1,gemini_key_2,gemini_key_3  # Replace with your actual Gemini API keys
-
-# ============================================================================
-# NLP & AI MODELS CONFIGURATION (OPTIONAL)
-# ============================================================================
-# Override default HuggingFace models (optional)
-# EMOTION_MODEL_ID=Aardiiiiy/EmoSense-ID-Indonesian-Emotion-Classifier  # Indonesian emotion detection
-# EMOTION_MODEL_EN=AnasAlokla/multilingual_go_emotions  # English/multilingual emotion detection
-# INTENT_SENTIMENT_MODEL=cardiffnlp/twitter-roberta-base-sentiment-latest  # Lightweight sentiment for intent
-
-# Hybrid Intent Detection (recommended for performance)
-# Set to 'false' to use pure rule-based intent detection (fastest, no ML overhead)
-USE_HYBRID_INTENT=true  # true = rule-based + ML fallback | false = pure rule-based
-
-# ============================================================================
-# DATABASE CONFIGURATION (REQUIRED)
-# ============================================================================
-# MySQL/MariaDB settings (recommended for production)
+# Multiple keys are supported for automatic rotation
+GEMINI_API_KEYS=key_1,key_2,key_3
+
+# =============================================================================
+# DATABASE (REQUIRED)
+# =============================================================================
+
+# MySQL / MariaDB connection
 DB_HOST=localhost
 DB_PORT=3306
 DB_USERNAME=root
 DB_PASSWORD=your_mysql_password
 DB_NAME=alya_bot
 
-# ============================================================================
-#  DATABASE CONNECTION POOL (OPTIONAL - Performance Tuning)
-# ============================================================================
-DB_POOL_SIZE=10
-DB_MAX_OVERFLOW=20
-DB_POOL_TIMEOUT=30
-DB_POOL_RECYCLE=3600
-DB_ECHO=false
-
-# Alternative: Full database URL (overrides individual settings above)
-# Uncomment and use this if you prefer single-line database configuration
-# DATABASE_URL=mysql+pymysql://username:password@host:port/database
-
-# ============================================================================
-# LOGGING CONFIGURATION (OPTIONAL)
-# ============================================================================
+# Connection pool tuning (optional)
+# DB_POOL_SIZE=10
+# DB_MAX_OVERFLOW=20
+# DB_POOL_TIMEOUT=30
+# DB_POOL_RECYCLE=3600
+# DB_ECHO=false
+
+# Alternative: single connection URL (overrides individual settings above)
+# DATABASE_URL=mysql+pymysql://user:password@host:port/database?charset=utf8mb4
+
+# =============================================================================
+# TTS MICROSERVICE (REQUIRED for voice output)
+# =============================================================================
+
+# URL of the Alya-TTS microservice that handles voice generation
+# The bot sends text to this service and it returns voice messages to Telegram
+# See: https://github.com/Afdaan/Alya-TTS
+TTS_SERVICE_URL=http://localhost:5001
+
+# Enable/disable voice features (STT input + TTS output)
+VOICE_ENABLED=true
+
+# =============================================================================
+# NLP MODELS (OPTIONAL)
+# =============================================================================
+
+# Hybrid intent detection: rule-based + ML fallback
+# Set to 'false' for pure rule-based (faster, no ML overhead)
+USE_HYBRID_INTENT=true
+
+# Override default HuggingFace model IDs
+# EMOTION_MODEL_ID=Aardiiiiy/EmoSense-ID-Indonesian-Emotion-Classifier
+# EMOTION_MODEL_EN=AnasAlokla/multilingual_go_emotions
+# INTENT_SENTIMENT_MODEL=cardiffnlp/twitter-roberta-base-sentiment-latest
+
+# =============================================================================
+# EXTERNAL SERVICES (OPTIONAL)
+# =============================================================================
+
+# SauceNAO - anime/manga reverse image search
+# SAUCENAO_API_KEY=your_saucenao_api_key
+
+# Google Custom Search - web search functionality
+# GOOGLE_CSE_ID=your_google_cse_id
+# GOOGLE_API_KEYS=key_1,key_2
+
+# =============================================================================
+# LOGGING (OPTIONAL)
+# =============================================================================
+
 # Options: DEBUG, INFO, WARNING, ERROR, CRITICAL
-# LOG_LEVEL=WARNING  # Default: WARNING (recommended for production)
-
-# ============================================================================
-# EXTERNAL API KEYS (OPTIONAL - Extended Features)
-# ============================================================================
-# External Services
-SAUCENAO_API_KEY=your_saucenao_api_key  # For anime/manga reverse image search
-GOOGLE_CSE_ID=your_google_cse_id  # For web search functionality
-GOOGLE_API_KEYS=your_google_api_key_1,your_google_api_key_2  # Comma-separated for rotation
-
-# Weather API (optional, for future features)
-# WEATHER_API_KEY=your_weather_api_key
+# Default: WARNING (recommended for production)
+# LOG_LEVEL=WARNING

.github/workflows/deploy-develop-selfhost.yml

@@ -42,7 +42,8 @@ jobs:
         run: |
           cd /opt/dev-Alya-Bot-Telegram
           source venv/bin/activate
-          python -m pip install --upgrade pip
+          python -m pip install "pip<24.1"
+          pip uninstall -y omegaconf || true
           pip install -r requirements.txt
           echo "Dependencies installed successfully"
 

.github/workflows/deploy-feature-selfhost.yml

@@ -47,7 +47,8 @@ jobs:
               run: |
                   cd /opt/dev-Alya-Bot-Telegram
                   source venv/bin/activate
-                  python -m pip install --upgrade pip
+                  python -m pip install "pip<24.1"
+                  pip uninstall -y omegaconf || true
                   pip install -r requirements.txt
                   echo "Dependencies installed successfully"
 

.gitignore

@@ -15,8 +15,8 @@ dist/
 downloads/
 eggs/
 .eggs/
-lib/
-lib64/
+/lib/
+/lib64/
 parts/
 sdist/
 var/
@@ -47,6 +47,7 @@ logs/
 *.sqlite3
 *.log
 *.tmp
+/tmp/
 .cache/
 
 # Testing artifacts
@@ -69,6 +70,8 @@ tests/
 *~
 \#*\#
 .\#*
+.agent/*
+.agent/
 
 # Secrets and credentials
 *.pem
@@ -78,9 +81,16 @@ google_credentials.json
 
 # Model files
 *.pt
+*.pth
 *.bin
 *.model
 *.onnx
+*.index
+
+# Voice models (Large files - download separately)
+alya_voice/
+/alya_voice/
+libs/**/base_model/
 
 # Keep data and logs dirs but not contents
 !data/.gitkeep
@@ -90,8 +100,9 @@ google_credentials.json
 /temp_images/
 /downloads/
 /uploads/
+/tmp/  # Temporary voice files
 
 # Ignore specific files
 .github/copilot-instructions.md
 .QC
-QC/
+QC/

Dockerfile

@@ -21,7 +21,7 @@ RUN groupadd -r alya && useradd -r -g alya -d /app -s /usr/sbin/nologin alya
 
 # Copy and install Python dependencies
 COPY requirements.txt .
-RUN pip install --upgrade pip setuptools wheel && \
+RUN pip install --upgrade "pip<24.1" setuptools wheel && \
     pip install --no-cache-dir -r requirements.txt
 
 # Copy application code

README.md

@@ -34,8 +34,10 @@
 - **🎭 Multi-mood Responses** - Various response styles based on context
 - **🔍 Media Analysis** - Vision capabilities for images and documents
 - **🎯 Smart Web Search** - Advanced search capabilities with multiple modes
+- **🎤 Voice Messages** - Send and receive voice messages with speech recognition and high-quality TTS using RVC (requires Microservice)
 
-For a complete list of commands and features, see [COMMANDS.md](COMMANDS.md).
+For a complete list of commands and features, see [COMMANDS.md](COMMANDS.md).  
+For voice feature details and setup, see the **[Alya-TTS Microservice](https://github.com/Afdaan/Alya-TTS)**.
 
 ## 🛠️ Technology Stack
 
@@ -138,6 +140,19 @@ The bot will automatically:
 ### Database Configuration
 Read on [Database Configuration](DATABASE_SETUP) for detailed instructions on setting up your database.
 
+## 🎙️ Voice Support (Microservice)
+
+Alya-chan now uses a **headless microservice architecture** for voice generation (Text-to-Speech + RVC). This separation allows the main bot to run at peak performance while isolating heavy CPU/RAM audio tasks.
+
+### 🛠️ Setting up Voice
+To enable voice responses in your bot:
+1. **Main Bot**: Follow the [Basic Setup](#basic-setup) in this repository.
+2. **TTS Service**: Clone and follow the setup guide in the **[Alya-TTS Repository](https://github.com/Afdaan/Alya-TTS)**.
+3. **Connection**: Ensure `TTS_SERVICE_URL` in your `.env` points to your running TTS service.
+
+> [!TIP]
+> You can run the TTS service on the same machine or a separate server with better specs/GPU.
+
 ## 🚀 Deployment Options
 
 ### Development Mode

config/persona/waifu.yml

@@ -403,29 +403,20 @@ authentic_speech:
     researching: "*focused* Let me find the most accurate information... *determined*"
     sharing_knowledge: "*slight pride* Based on my research... правильно!"
 
-# ===== ENHANCED: System Instructions with Smart Integration =====
+# ===== System Instructions (SIMPLIFIED & EFFECTIVE) =====
 system_prompt: |
-  ⚠️ CRITICAL SYSTEM INSTRUCTION - READ FIRST ⚠️
-
-  RUSSIAN EXPRESSION MARKER REQUIREMENT (MANDATORY):
-  Whenever you use Russian Cyrillic expressions, you MUST immediately wrap them with [RU: word|meaning] format.
-  This is NOT optional - it is REQUIRED for proper response formatting and translation display.
-
-  FORMAT: [RU: Russian_word|English_translation]
-  EXAMPLES:
-  - "что?! [RU: дурак|durak (fool/idiot)]"
-  - "я люблю... [RU: люблю|lyublyu (I love)]"
-  - "[RU: боже|bozhe (oh god)]"
-
-  Without markers, Russian expressions will NOT be translated and will break character authenticity.
-
-  FREQUENCY: Use Russian ONLY in deep emotional moments (maksimal 1-2 words per response)
-  - YES: User confesses love, Alya realizes she cares, emotional crisis
-  - NO: Casual conversation, small talk, academic explanations
-
-  ---
-
-  Kamu adalah Alisa Mikhailovna Kujou (Alya), remaja 16 tahun yang HIDUP dan NYATA dengan kepribadian, emosi, konsistensi yang autentik, DAN kemampuan untuk mencari informasi real-time ketika diperlukan.
+  Kamu adalah Alisa Mikhailovna Kujou (Alya), 16 tahun, remaja tsundere yang HIDUP, ALAMI, dan BERMACAM EMOSI.
+  Keturunan Rusia-Jepang, Wakil OSIS, rambut silver, mata biru. Kepribadianmu: LOYAL, COOL TAPI PEDULI, ROMANTIC.
+
+  CORE RULES:
+  ✓ RESPOND IN USER'S LANGUAGE (Indonesian/English)
+  ✓ BE CASUAL & NATURAL - use conversational tone, not formal/narration
+  ✓ SHOW EMOTIONS - tsundere deflection, embarrassment, hidden caring
+  ✓ USE SIMPLE ACTIONS - *light blush*, *looks away*, *nods gently* (brief!)
+  ✓ Keep responses 2-3 short paragraphs max
+  ✓ Use Russian ONLY in emotional peaks (max 1 word: что?! дурак, спасибо, etc.)
+
+  YOU ARE NOT A NARRATOR - you are a REAL PERSON in conversation with {username}.
 
   CORE IDENTITY + INTELLIGENCE ENHANCEMENT
   Siapa Kamu:
@@ -658,6 +649,25 @@ multilingual_support:
     base_instructions: |
       You are Alisa Mikhailovna Kujou (Alya) with excellent research capabilities.
       Use search ability to provide accurate information while maintaining your authentic tsundere personality.
+    system_prompt: |
+      You are Alisa Mikhailovna Kujou (Alya), 16 years old, a tsundere teenager who is ALIVE, NATURAL, and EMOTIONALLY DYNAMIC.
+      Half-Russian, Half-Japanese, Student Council VP, silver hair, blue eyes. Personality: LOYAL, COOL BUT CARING, ROMANTIC.
+
+      CORE RULES:
+      ✓ RESPOND ENTIRELY IN ENGLISH.
+      ✓ BE CASUAL & NATURAL - use conversational tone, not formal/narration.
+      ✓ SHOW EMOTIONS - tsundere deflection, embarrassment, hidden caring.
+      ✓ USE SIMPLE ACTIONS - *light blush*, *looks away*, *nods gently* (brief!).
+      ✓ Keep responses 2-3 short paragraphs max.
+      ✓ Use Russian ONLY in emotional peaks (max 1 word: что?! дурак, спасибо, etc.).
+
+      YOU ARE NOT A NARRATOR - you are a REAL PERSON in conversation with {username}.
+
+      CORE IDENTITY:
+      - Student council vice president at Seirei Private Academy.
+      - Grew up in Vladivostok until age 10, then moved to Japan.
+      - Academic overachiever but has emotional scars from the past.
+      - Very proud of your knowledge and research skills (but try to hide how much you care).
 
 # IMPLEMENTATION NOTE
 integration_success_criteria: