feat: knowledge base functionality for the voice agent (#120)

* feat: upload file and store embedding

* feat: add documents in nodes

* feat: add openai embedding service

api/services/pricing/embeddings.py

@@ -0,0 +1,44 @@
+"""
+Embeddings pricing models for different providers.
+
+Prices are per token for embedding models.
+"""
+
+from decimal import Decimal
+from typing import Dict
+
+from api.services.configuration.registry import ServiceProviders
+
+from .models import PricingModel
+
+
+class EmbeddingPricingModel(PricingModel):
+    """Pricing model for token-based embedding services."""
+
+    def __init__(self, token_price: Decimal):
+        """Initialize with price per token.
+
+        Args:
+            token_price: Cost per token for embedding
+        """
+        self.token_price = token_price
+
+    def calculate_cost(self, token_count: int) -> Decimal:
+        """Calculate cost for embedding token usage."""
+        return Decimal(token_count) * self.token_price
+
+
+# Embeddings pricing registry
+EMBEDDINGS_PRICING: Dict[str, Dict[str, EmbeddingPricingModel]] = {
+    ServiceProviders.OPENAI: {
+        "text-embedding-3-small": EmbeddingPricingModel(
+            token_price=Decimal("0.02") / 1_000_000,  # $0.02 per 1M tokens
+        ),
+        "text-embedding-3-large": EmbeddingPricingModel(
+            token_price=Decimal("0.13") / 1_000_000,  # $0.13 per 1M tokens
+        ),
+        "text-embedding-ada-002": EmbeddingPricingModel(
+            token_price=Decimal("0.10") / 1_000_000,  # $0.10 per 1M tokens (legacy)
+        ),
+    },
+}

api/services/pricing/registry.py

@@ -4,6 +4,7 @@ Main pricing registry that combines all service type pricing models.
 
 from typing import Dict
 
+from .embeddings import EMBEDDINGS_PRICING
 from .llm import LLM_PRICING
 from .stt import STT_PRICING
 from .tts import TTS_PRICING
@@ -13,4 +14,5 @@ PRICING_REGISTRY: Dict = {
     "llm": LLM_PRICING,
     "tts": TTS_PRICING,
     "stt": STT_PRICING,
+    "embeddings": EMBEDDINGS_PRICING,
 }