chore: sync sdk code with DeepLearning repo (#142)

Co-authored-by: AssemblyAI <engineering.sdk@assemblyai.com>

Author: he-james

assemblyai/__version__.py

@@ -1 +1 @@
-__version__ = "0.45.1"
+__version__ = "0.45.2"

assemblyai/types.py

@@ -668,6 +668,9 @@ class RawTranscriptionConfig(BaseModel):
     keyterms_prompt: Optional[List[str]] = None
     "The list of key terms used to generate the transcript with the Slam-1 speech model. Can't be used together with `prompt`."
 
+    language_codes: Optional[List[Union[str, LanguageCode]]] = None
+    "List of language codes detected in the audio file when language detection is enabled"
+
     model_config = ConfigDict(extra="allow")
 
 
@@ -1237,6 +1240,12 @@ def speech_threshold(self, threshold: Optional[float]) -> None:
 
         self._raw_transcription_config.speech_threshold = threshold
 
+    @property
+    def language_codes(self) -> Optional[List[Union[str, LanguageCode]]]:
+        "Returns the list of language codes detected in the audio file when language detection is enabled."
+
+        return self._raw_transcription_config.language_codes
+
     # endregion
 
     # region: Convenience (helper) methods
@@ -1913,6 +1922,9 @@ class BaseTranscript(BaseModel):
     language_confidence: Optional[float] = None
     "The confidence score for the detected language, between 0.0 (low confidence) and 1.0 (high confidence)."
 
+    language_codes: Optional[List[Union[str, LanguageCode]]] = None
+    "List of language codes detected in the audio file when language detection is enabled"
+
     speech_threshold: Optional[float] = None
     "Reject audio files that contain less than this fraction of speech. Valid values are in the range [0,1] inclusive"
 

tests/unit/test_transcript.py

@@ -452,6 +452,17 @@ def test_delete_by_id_async(httpx_mock: HTTPXMock):
     assert transcript.audio_url == mock_transcript_response["audio_url"]
 
 
+def test_language_codes_field():
+    """Test language_codes field deserialization"""
+    mock_response = factories.generate_dict_factory(
+        factories.TranscriptCompletedResponseFactory
+    )()
+    mock_response["language_codes"] = ["en", "es"]
+
+    response = aai.types.TranscriptResponse(**mock_response)
+    assert response.language_codes == ["en", "es"]
+
+
 def test_speech_model_used_field_deserialization():
     """
     Tests that the speech_model_used field can be properly deserialized.