add language detection for AWS Transcribe Service

Author: kennethtegrado

src/pipecat/services/aws/__init__.py

@@ -8,6 +8,7 @@
 
 from pipecat.services import DeprecatedModuleProxy
 
+from .config import *
 from .llm import *
 from .nova_sonic import *
 from .stt import *

src/pipecat/services/aws/config.py

@@ -0,0 +1,53 @@
+#
+# Copyright (c) 2024–2025, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+"""Configuration for the AWS Transcribe STT service."""
+
+from typing import List, Optional
+
+from pydantic import BaseModel
+
+from pipecat.transcriptions.language import Language
+
+
+class AWSInputParams(BaseModel):
+    """Configuration parameters for the AWS Transcribe STT service.
+
+    Parameters:
+        sample_rate: Audio sample rate in Hz. Must be 8000 or 16000. Defaults to 16000.
+        language_code: Language for transcription. Cannot be used with identify_multiple_languages.
+        language_options: List of languages for multi-language identification. Required when identify_multiple_languages is True.
+        identify_multiple_languages: Enable multiple language identification. Defaults to False.
+        identify_language: Enable language identification. Defaults to False.
+        preferred_language: Preferred language from language_options to speed up identification.
+        enable_partial_results_stabilization: Enable stabilization of partial results. Defaults to True.
+        partial_results_stability: Stability level: "low" (faster, less stable) or "high" (slower, more stable). Defaults to "low".
+        media_encoding: Audio encoding format. Defaults to "linear16".
+        number_of_channels: Number of audio channels. Defaults to 1.
+        show_speaker_label: Enable speaker identification in transcription results. Defaults to False.
+        enable_channel_identification: Enable channel identification for multi-channel audio. Defaults to False.
+        vocabulary_name: Name of custom vocabulary to use for improved transcription accuracy.
+        vocabulary_filter_name: Name of vocabulary filter to apply for content filtering.
+
+    Note:
+        For real-time conversations, use partial_results_stability="low" for faster responses.
+        Multi-language identification may have higher latency than single language mode.
+    """
+
+    sample_rate: int = 16000
+    language_code: Optional[Language] = None
+    language_options: Optional[List[Language]] = None
+    identify_multiple_languages: bool = False
+    identify_language: bool = False
+    preferred_language: Optional[Language] = None
+    enable_partial_results_stabilization: bool = True
+    partial_results_stability: str = "high"
+    media_encoding: str = "linear16"
+    number_of_channels: int = 1
+    show_speaker_label: bool = False
+    enable_channel_identification: bool = False
+    vocabulary_name: Optional[str] = None
+    vocabulary_filter_name: Optional[str] = None