oip-arweave-indexer/docker-compose-backend-only.yml at main · DevonJames/oip-arweave-indexer · GitHub

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
version: '3.8'

services:
  # Main application (backend only - no client-side services)
  main-app:
    build: .
    ports:
      - "3000:3000"
    volumes:
      - ./uploads:/app/uploads
      - ./logs:/app/logs
      - models:/app/models
    environment:
      # Backend-only configuration
      - DISTRIBUTED_MODE=true
      - CLIENT_SERVICES_EXTERNAL=true

      # Elasticsearch configuration
      - ELASTICSEARCH_URL=http://elasticsearch:9200
      - ELASTICSEARCH_INDEX=oip-records

      # Ollama configuration
      - OLLAMA_URL=http://ollama:11434
      - OLLAMA_MODEL=llama3.1:8b

      # Enhanced TTS configuration (Kokoro + fallbacks)
      - TTS_SERVICE_URL=http://kokoro-tts:8012
      - TTS_PRIMARY_ENGINE=kokoro
      - TTS_FALLBACK_ENGINES=coqui,piper,espeak
      - CACHE_ENABLED=true
      - DEFAULT_VOICE=default
      - SAMPLE_RATE=22050
      - AUDIO_FORMAT=wav
      - OFFLINE_MODE=false
      - DISABLE_GTTS=false
      - ENABLE_CHATTERBOX=true
      - ENABLE_ESPEAK=true

      # External client service URLs (Mac client)
      - STT_SERVICE_URL=http://mac-client:8013
      - SMART_TURN_URL=http://mac-client:8014
      - STT_ENABLED=false  # Handled by external Mac client
      - SMART_TURN_ENABLED=false  # Handled by external Mac client
      - VAD_ENABLED=false  # Handled by external Mac client

      # Pipeline configuration
      - ENHANCED_PIPELINE_ENABLED=true
      - PIPELINE_VERSION=2.0-distributed

    depends_on:
      - elasticsearch
      - ollama
      - kokoro-tts
    networks:
      - voice-network

  # Elasticsearch for document storage and search
  elasticsearch:
    image: docker.elastic.co/elasticsearch/elasticsearch:8.11.0
    environment:
      - discovery.type=single-node
      - xpack.security.enabled=false
      - "ES_JAVA_OPTS=-Xms2g -Xmx2g"
    ports:
      - "9200:9200"
    volumes:
      # Bind mount to host filesystem instead of Docker-managed volume
      # This prevents hitting Docker's disk image space limits
      # Set ELASTICSEARCH_DATA_PATH in .env to customize location (default: ./elasticsearch_data)
      - ${ELASTICSEARCH_DATA_PATH:-./elasticsearch_data}:/usr/share/elasticsearch/data
    networks:
      - voice-network

  # Ollama for LLM processing
  ollama:
    image: ollama/ollama:latest
    ports:
      - "11434:11434"
    volumes:
      - ollama_data:/root/.ollama
    environment:
      - NVIDIA_VISIBLE_DEVICES=all
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    networks:
      - voice-network

  # Kokoro TTS Service (primary TTS engine with fallbacks)
  kokoro-tts:
    build: ./kokoro-tts-service
    ports:
      - "8012:8012"
    volumes:
      - models:/app/models
      - ./logs:/app/logs
    environment:
      # Kokoro TTS configuration
      - TTS_PRIMARY_ENGINE=kokoro
      - TTS_FALLBACK_ENGINES=coqui,piper,espeak
      - MODEL_PATH=/app/models/kokoro
      - CACHE_ENABLED=true
      - DEFAULT_VOICE=default
      - SAMPLE_RATE=22050
      - AUDIO_FORMAT=wav
      - OFFLINE_MODE=false

      # Fallback engine configuration
      - COQUI_MODEL=tts_models/en/ljspeech/tacotron2-DDC
      - PIPER_MODEL=en_US-lessac-medium
      - ESPEAK_VOICE=en
      - DISABLE_GTTS=false
      - ENABLE_CHATTERBOX=true
      - ENABLE_ESPEAK=true

      # Performance settings
      - MAX_TEXT_LENGTH=1000
      - CONCURRENT_REQUESTS=4
      - REQUEST_TIMEOUT=30
      - ENABLE_STREAMING=false

    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    networks:
      - voice-network

volumes:
  # Note: elasticsearch_data removed - now using bind mount to host filesystem
  # This prevents hitting Docker's disk image space limits
  ollama_data:
  models:

networks:
  voice-network:
    driver: bridge
    ipam:
      config:
        - subnet: 172.20.0.0/16