omi/backend/pusher/requirements.txt at main · ghzhost/omi · 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
# Pusher service requirements — lightweight subset of backend/requirements.txt
# No torch, onnxruntime, pyannote, speechbrain, librosa, matplotlib, pandas, scipy

# Web framework
fastapi==0.121.0
uvicorn==0.30.5
uvloop==0.20.0
starlette==0.49.3
websockets==12.0
httpx==0.28.0
python-multipart==0.0.31

# Data validation
pydantic==2.11.10
pydantic-settings==2.10.1
pydantic_core==2.33.2
jsonschema==4.23.0
jsonschema-specifications==2023.12.1

# Google Cloud
firebase-admin==6.5.0
google-cloud-firestore==2.20.0
google-cloud-storage==2.18.0
google-cloud-tasks==2.16.4
google-auth==2.32.0
google-api-python-client==2.139.0

# Cache & rate limiting
redis==5.0.8

# HTTP
requests~=2.33.0

# Audio encoding & processing
opuslib==3.0.1
av==12.0.0
pydub==0.25.1
lc3py==1.1.3
webrtcvad==2.0.10

# Numeric
numpy==1.26.4

# Metrics
prometheus-client==0.21.1

# LLM services (for process_conversation)
anthropic>=0.52.0
openai==2.38.0
groq==0.9.0
langchain==1.2.10
langchain-anthropic==1.1.0
langchain-community==0.4.1
langchain-core==1.3.3
langchain-google-genai==3.2.0
langchain-openai==1.1.14
langchain-pinecone==0.2.13
langchain-text-splitters==1.1.2
langgraph==1.0.10
langsmith==0.8.5

# Vector databases
pinecone==7.3.0
qdrant-client==1.11.0

# Knowledge graph
neo4j==5.23.1

# Speech-to-text (Deepgram SDK for transcript handling)
deepgram-sdk==4.8.1

# Security & encryption
cryptography==48.0.1

# Image processing (via tools/__init__ -> file_tools -> chat_file -> PIL)
pillow==12.3.0

# Utilities
python-dotenv==1.2.2
tenacity==8.2.3
python-dateutil==2.9.0.post0
orjson==3.11.6
tiktoken==0.7.0
jiwer==3.0.4
python-ulid==3.0.0
pytz==2024.1

# Payment & communication
stripe==11.3.0
twilio==9.5.0

# Locale
pycountry==24.6.1
langdetect==1.0.9

# Speaker embedding (via speaker_identification -> speaker_embedding -> scipy)
scipy==1.14.0