Add model field to TranscriptionConfig for batch and RT

sdk/batch/README.md

@@ -118,7 +118,7 @@ async def main():
             type=JobType.TRANSCRIPTION,
             transcription_config=TranscriptionConfig(
                 language="en",
-                operating_point=OperatingPoint.ENHANCED,
+                model=Model.ENHANCED,
                 enable_entities=True,
                 diarization="speaker",
             ),

sdk/batch/pyproject.toml

@@ -10,7 +10,7 @@ readme = "README.md"
 authors = [{ name = "Speechmatics", email = "support@speechmatics.com" }]
 license = "MIT"
 requires-python = ">=3.9"
-dependencies = ["aiohttp", "aiofiles"]
+dependencies = ["aiohttp", "aiofiles", "typing-extensions>=4.5.0"]
 classifiers = [
     "Development Status :: 4 - Beta",
     "Intended Audience :: Developers",

sdk/batch/speechmatics/batch/_async_client.py

@@ -8,6 +8,7 @@
 from __future__ import annotations
 
 import asyncio
+import logging
 import os
 import uuid
 from typing import Any
@@ -191,6 +192,11 @@ async def submit_job(
             transcription_config = transcription_config or TranscriptionConfig()
             config = JobConfig(type=JobType.TRANSCRIPTION, transcription_config=transcription_config)
 
+        if config.transcription_config is not None and config.transcription_config.operating_point is not None:
+            logging.warning(
+                "TranscriptionConfig.operating_point is deprecated and will be removed in the future. Please use the model property instead."
+            )
+
         # Check for fetch_data configuration
         config_dict = config.to_dict()
         has_fetch_data = "fetch_data" in config_dict

sdk/batch/speechmatics/batch/_models.py

@@ -14,6 +14,10 @@
 from enum import Enum
 from typing import Any
 from typing import Optional
+from typing import cast
+from warnings import warn
+
+from typing_extensions import deprecated
 
 
 class JobType(str, Enum):
@@ -38,13 +42,21 @@ class JobStatus(str, Enum):
     EXPIRED = "expired"
 
 
+@deprecated("Use Model instead")
 class OperatingPoint(str, Enum):
     """Operating point options for transcription."""
 
     ENHANCED = "enhanced"
     STANDARD = "standard"
 
 
+class Model(str, Enum):
+    """Operating point options for transcription."""
+
+    ENHANCED = "enhanced"
+    STANDARD = "standard"
+
+
 class NotificationContents(str, Enum):
     """Notification content options."""
 
@@ -75,6 +87,9 @@ class FormatType(str, Enum):
     SRT = "srt"
 
 
+_UNSET = cast(Model, object())
+
+
 @dataclass
 class TranscriptionConfig:
     """
@@ -83,7 +98,7 @@ class TranscriptionConfig:
     Attributes:
         language: ISO 639-1 language code (e.g., "en", "es", "fr").
             defaults to "en"
-        operating_point: Which acoustic model to use.
+        model: Which acoustic model to use.
             defaults to "enhanced"
         output_locale: RFC-5646 language code for transcript output.
         diarization: Type of diarization to use. Options: "none", "speaker".
@@ -103,7 +118,7 @@ class TranscriptionConfig:
     """
 
     language: str = "en"
-    operating_point: OperatingPoint = OperatingPoint.ENHANCED
+    model: Model = _UNSET
     output_locale: Optional[str] = None
     diarization: Optional[str] = None
     additional_vocab: Optional[list[dict[str, Any]]] = None
@@ -117,6 +132,19 @@ class TranscriptionConfig:
     max_delay_mode: Optional[str] = None
     transcript_filtering_config: Optional[TranscriptFilteringConfig] = None
     audio_filtering_config: Optional[AudioFilteringConfig] = None
+    operating_point: Optional[OperatingPoint] = None
+
+    def __post_init__(self) -> None:
+        if self.model is not _UNSET and self.operating_point is not None:
+            raise ValueError("Cannot specify both 'model' and 'operating_point'. Use 'model' instead.")
+        if self.model is _UNSET and self.operating_point is None:
+            self.model = Model.ENHANCED
+        if self.operating_point is not None:
+            warn(
+                "'operating_point' is deprecated, use 'model' instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
 
     def to_dict(self) -> dict[str, Any]:
         result: dict[str, Any] = {k: v for k, v in asdict(self).items() if v is not None}

sdk/rt/speechmatics/rt/_async_client.py

@@ -23,6 +23,7 @@
 
 _UNSET = object()
 
+
 class AsyncClient(_BaseClient):
     """
     Asynchronous client for Speechmatics real-time audio transcription.
@@ -195,7 +196,7 @@ async def force_end_of_utterance(self, *, timestamp: Optional[float] | object =
                 ...     await client.force_end_of_utterance()
         """
 
-        message: dict[str,Any] = {"message": ClientMessageType.FORCE_END_OF_UTTERANCE}
+        message: dict[str, Any] = {"message": ClientMessageType.FORCE_END_OF_UTTERANCE}
 
         if timestamp is _UNSET:
             # default: auto-set from audio_seconds_sent

sdk/rt/speechmatics/rt/_base_client.py

@@ -3,6 +3,7 @@
 import asyncio
 import contextlib
 import json
+import logging
 import os
 import uuid
 from typing import Any
@@ -194,6 +195,11 @@ async def _start_recognition_session(
         transcription_config = transcription_config or TranscriptionConfig()
         audio_format = audio_format or AudioFormat()
 
+        if transcription_config.operating_point is not None:
+            logging.warning(
+                "TranscriptionConfig.operating_point is deprecated and will be removed in the future. Please use the model property instead."
+            )
+
         start_recognition_message = build_start_recognition_message(
             transcription_config=transcription_config,
             audio_format=audio_format,

sdk/rt/speechmatics/rt/_models.py

@@ -7,6 +7,9 @@
 from enum import Enum
 from typing import Any
 from typing import Optional
+from typing import cast
+from warnings import deprecated
+from warnings import warn
 
 
 class AudioEncoding(str, Enum):
@@ -30,13 +33,21 @@ class AudioEncoding(str, Enum):
     MULAW = "mulaw"
 
 
+@deprecated("Use Model instead")
 class OperatingPoint(str, Enum):
     """Operating point options for transcription."""
 
     ENHANCED = "enhanced"
     STANDARD = "standard"
 
 
+class Model(str, Enum):
+    """Which model to use for transcription."""
+
+    ENHANCED = "enhanced"
+    STANDARD = "standard"
+
+
 @dataclass
 class AudioEventsConfig:
     types: Optional[list[str]] = None
@@ -329,6 +340,9 @@ class SpeakerIdentifier:
     speaker_identifiers: list[str] = field(default_factory=list)
 
 
+_UNSET = cast(Model, object())
+
+
 @dataclass
 class TranscriptionConfig:
     """
@@ -337,7 +351,7 @@ class TranscriptionConfig:
     Attributes:
         language: (Optional) ISO 639-1 language code (e.g., "en", "es", "fr").
             Defaults to "en".
-        operating_point: (Optional) Which acoustic model to use.
+        model: (Optional) Which acoustic model to use.
             Defaults to "enhanced".
         output_locale: (Optional) RFC-5646 language code for transcript output (eg. "en-US").
             Defaults to None.
@@ -373,7 +387,8 @@ class TranscriptionConfig:
             Defaults to None.
         channel_diarization_labels: (Optional) Configuration for channel diarization.
             Defaults to None.
-
+        operating_point: (Deprecated) Legacy argument for specifying the operating point. Use `model` instead going forward
+            Defaults to None.
 
     Examples:
         Basic English transcription:
@@ -382,7 +397,7 @@ class TranscriptionConfig:
         Spanish with partials enabled:
             >>> config = TranscriptionConfig(
             ...     language="es",
-            ...     operating_point="enhanced",
+            ...     model="enhanced",
             ...     enable_partials=True
             ... )
 
@@ -399,7 +414,7 @@ class TranscriptionConfig:
     """
 
     language: str = "en"
-    operating_point: OperatingPoint = OperatingPoint.ENHANCED
+    model: Model = _UNSET
     output_locale: Optional[str] = None
     diarization: Optional[str] = None
     additional_vocab: Optional[list[dict[str, Any]]] = None
@@ -416,6 +431,19 @@ class TranscriptionConfig:
     conversation_config: Optional[ConversationConfig] = None
     ctrl: Optional[dict] = None
     channel_diarization_labels: Optional[list[str]] = None
+    operating_point: Optional[OperatingPoint] = None
+
+    def __post_init__(self) -> None:
+        if self.model is not _UNSET and self.operating_point is not None:
+            raise ValueError("Cannot specify both 'model' and 'operating_point'. Use 'model' instead.")
+        if self.model is _UNSET and self.operating_point is None:
+            self.model = Model.ENHANCED
+        if self.operating_point is not None:
+            warn(
+                "'operating_point' is deprecated, use 'model' instead.",
+                DeprecationWarning,
+                stacklevel=2,
+            )
 
     def to_dict(self) -> dict[str, Any]:
         """