feat: add Azure AI multi-provider support (TTS, STT, Embeddings, Realtime)

Enables Azure AI services across all model layers so users with Azure
credits can consolidate billing on a single provider.

- Voice (TTS): AzureSpeechTTSConfiguration via azure_speech provider
- Transcriber (STT): AzureSpeechSTTConfiguration via azure_speech provider
- Embedding: AzureOpenAIEmbeddingsConfiguration via azure provider
- Realtime: AzureRealtimeLLMConfiguration via azure_realtime provider

New files:
- api/services/pipecat/realtime/azure_realtime.py
- api/services/gen_ai/embedding/azure_openai_service.py
- api/tests/test_azure_speech_service_factory.py

The UI picks up all four providers automatically from the schema —
no frontend changes required.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

api/tests/test_azure_speech_service_factory.py

@@ -0,0 +1,81 @@
+"""Tests for Azure Speech TTS/STT service factory dispatch."""
+
+from types import SimpleNamespace
+from unittest.mock import patch
+
+from api.services.configuration.registry import ServiceProviders
+from api.services.pipecat.service_factory import create_stt_service, create_tts_service
+
+
+def _audio_config():
+    return SimpleNamespace(
+        transport_out_sample_rate=24000,
+        transport_in_sample_rate=16000,
+    )
+
+
+def test_create_azure_speech_tts_service():
+    user_config = SimpleNamespace(
+        tts=SimpleNamespace(
+            provider=ServiceProviders.AZURE_SPEECH.value,
+            api_key="test-subscription-key",
+            region="eastus",
+            voice="en-US-AriaNeural",
+            language="en-US",
+            speed=1.0,
+            model="neural",
+        )
+    )
+
+    with patch("api.services.pipecat.service_factory.AzureTTSService") as mock_service:
+        create_tts_service(user_config, _audio_config())
+
+    assert mock_service.call_count == 1
+    kwargs = mock_service.call_args.kwargs
+    assert kwargs["api_key"] == "test-subscription-key"
+    assert kwargs["region"] == "eastus"
+    assert kwargs["settings"].voice == "en-US-AriaNeural"
+    assert kwargs["settings"].language == "en-US"
+
+
+def test_create_azure_speech_tts_service_with_speed():
+    user_config = SimpleNamespace(
+        tts=SimpleNamespace(
+            provider=ServiceProviders.AZURE_SPEECH.value,
+            api_key="test-key",
+            region="westeurope",
+            voice="en-GB-SoniaNeural",
+            language="en-GB",
+            speed=1.5,
+            model="neural",
+        )
+    )
+
+    with patch("api.services.pipecat.service_factory.AzureTTSService") as mock_service:
+        create_tts_service(user_config, _audio_config())
+
+    assert mock_service.call_count == 1
+    kwargs = mock_service.call_args.kwargs
+    assert kwargs["region"] == "westeurope"
+    assert kwargs["settings"].rate == "1.5"
+
+
+def test_create_azure_speech_stt_service():
+    user_config = SimpleNamespace(
+        stt=SimpleNamespace(
+            provider=ServiceProviders.AZURE_SPEECH.value,
+            api_key="test-subscription-key",
+            region="eastus",
+            language="en-US",
+            model="latest_long",
+        )
+    )
+
+    with patch("api.services.pipecat.service_factory.AzureSTTService") as mock_service:
+        create_stt_service(user_config, _audio_config())
+
+    assert mock_service.call_count == 1
+    kwargs = mock_service.call_args.kwargs
+    assert kwargs["api_key"] == "test-subscription-key"
+    assert kwargs["region"] == "eastus"
+    assert kwargs["sample_rate"] == 16000