cartesia 2.0.4__py3-none-any.whl → 2.0.6__py3-none-any.whl

cartesia/stt/socket_client.py

@@ -0,0 +1,138 @@
+import typing
+from typing import Any, Dict, Generator, Optional
+
+from ..core.client_wrapper import AsyncClientWrapper, SyncClientWrapper
+from ._async_websocket import AsyncSttWebsocket
+from ._websocket import SttWebsocket
+from .client import AsyncSttClient, SttClient
+from .types.stt_encoding import SttEncoding
+
+
+class SttClientWithWebsocket(SttClient):
+    """
+    Extension of STT functionality that supports a synchronous WebSocket STT connection.
+    """
+
+    def __init__(self, *, client_wrapper: SyncClientWrapper):
+        super().__init__(client_wrapper=client_wrapper)
+
+    def _ws_url(self):
+        base_url = self._client_wrapper.get_base_url()
+        if base_url.startswith("ws://") or base_url.startswith("wss://"):
+            return base_url
+        else:
+            prefix = "ws" if "localhost" in base_url else "wss"
+            base_url_without_protocol = base_url.split("://")[-1]
+            return f"{prefix}://{base_url_without_protocol}"
+
+    def websocket(
+        self,
+        *,
+        model: str = "ink-whisper",
+        language: Optional[str] = "en", 
+        encoding: SttEncoding = "pcm_s16le",
+        sample_rate: int = 16000,
+        min_volume: Optional[float] = None,
+        max_silence_duration_secs: Optional[float] = None,
+    ):
+        """Create a WebSocket connection for real-time speech transcription.
+
+        Args:
+            model: ID of the model to use for transcription
+            language: The language of the input audio in ISO-639-1 format
+            encoding: The encoding format of the audio data (required)
+            sample_rate: The sample rate of the audio in Hz (required)
+            min_volume: Volume threshold for voice activity detection (0.0-1.0)
+            max_silence_duration_secs: Maximum duration of silence before endpointing
+
+        Returns:
+            SttWebsocket: A connected WebSocket client for STT operations.
+            
+        Example:
+            >>> client = Cartesia(api_key="your-api-key")
+            >>> ws = client.stt.websocket()
+            >>> for result in ws.transcribe(audio_chunks):
+            ...     print(result["text"])
+        """
+        client_headers = self._client_wrapper.get_headers()
+        ws = SttWebsocket(
+            ws_url=self._ws_url(),
+            cartesia_version=client_headers["Cartesia-Version"],
+            api_key=client_headers["X-API-Key"],
+        )
+        # Auto-connect like TTS does for consistency
+        ws.connect(
+            model=model,
+            language=language,
+            encoding=encoding,
+            sample_rate=sample_rate,
+            min_volume=min_volume,
+            max_silence_duration_secs=max_silence_duration_secs,
+        )
+        return ws
+
+
+class AsyncSttClientWithWebsocket(AsyncSttClient):
+    """
+    Extension of STT functionality that supports an asynchronous WebSocket STT connection.
+    """
+
+    def __init__(self, *, client_wrapper: AsyncClientWrapper, get_session):
+        super().__init__(client_wrapper=client_wrapper)
+        self._get_session = get_session
+
+    def _ws_url(self) -> str:
+        base_url = self._client_wrapper.get_base_url()
+        if base_url.startswith("ws://") or base_url.startswith("wss://"):
+            return base_url
+        else:
+            prefix = "ws" if "localhost" in base_url else "wss"
+            base_url_without_protocol = base_url.split("://")[-1]
+            return f"{prefix}://{base_url_without_protocol}"
+
+    async def websocket(
+        self,
+        *,
+        model: str = "ink-whisper",
+        language: Optional[str] = "en",
+        encoding: SttEncoding = "pcm_s16le", 
+        sample_rate: int = 16000,
+        min_volume: Optional[float] = None,
+        max_silence_duration_secs: Optional[float] = None,
+    ):
+        """Create an async WebSocket connection for real-time speech transcription.
+
+        Args:
+            model: ID of the model to use for transcription
+            language: The language of the input audio in ISO-639-1 format
+            encoding: The encoding format of the audio data (required)
+            sample_rate: The sample rate of the audio in Hz (required)
+            min_volume: Volume threshold for voice activity detection (0.0-1.0)
+            max_silence_duration_secs: Maximum duration of silence before endpointing
+
+        Returns:
+            AsyncSttWebsocket: A connected async WebSocket client for STT operations.
+            
+        Example:
+            >>> client = AsyncCartesia(api_key="your-api-key")
+            >>> ws = await client.stt.websocket()
+            >>> async for result in ws.transcribe(audio_chunks):
+            ...     print(result["text"])
+        """
+        client_headers = self._client_wrapper.get_headers()
+        ws = AsyncSttWebsocket(
+            ws_url=self._ws_url(),
+            cartesia_version=client_headers["Cartesia-Version"],
+            api_key=client_headers["X-API-Key"],
+            get_session=self._get_session,
+        )
+        # Auto-connect like TTS does for consistency
+        await ws.connect(
+            model=model,
+            language=language,
+            encoding=encoding,
+            sample_rate=sample_rate,
+            min_volume=min_volume,
+            max_silence_duration_secs=max_silence_duration_secs,
+        )
+        return ws 

cartesia/stt/types/__init__.py

@@ -0,0 +1,33 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from .done_message import DoneMessage
+from .error_message import ErrorMessage
+from .flush_done_message import FlushDoneMessage
+from .streaming_transcription_response import (
+    StreamingTranscriptionResponse,
+    StreamingTranscriptionResponse_Done,
+    StreamingTranscriptionResponse_Error,
+    StreamingTranscriptionResponse_FlushDone,
+    StreamingTranscriptionResponse_Transcript,
+)
+from .stt_encoding import SttEncoding
+from .timestamp_granularity import TimestampGranularity
+from .transcript_message import TranscriptMessage
+from .transcription_response import TranscriptionResponse
+from .transcription_word import TranscriptionWord
+
+__all__ = [
+    "DoneMessage",
+    "ErrorMessage",
+    "FlushDoneMessage",
+    "StreamingTranscriptionResponse",
+    "StreamingTranscriptionResponse_Done",
+    "StreamingTranscriptionResponse_Error",
+    "StreamingTranscriptionResponse_FlushDone",
+    "StreamingTranscriptionResponse_Transcript",
+    "SttEncoding",
+    "TimestampGranularity",
+    "TranscriptMessage",
+    "TranscriptionResponse",
+    "TranscriptionWord",
+]

cartesia/stt/types/done_message.py

@@ -0,0 +1,26 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ...core.pydantic_utilities import UniversalBaseModel
+import pydantic
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+import typing
+
+
+class DoneMessage(UniversalBaseModel):
+    """
+    Acknowledgment message sent in response to a `done` command, indicating that the session is complete and the WebSocket will close.
+    """
+
+    request_id: str = pydantic.Field()
+    """
+    Unique identifier for this transcription session.
+    """
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow

cartesia/stt/types/error_message.py

@@ -0,0 +1,27 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ...core.pydantic_utilities import UniversalBaseModel
+import typing
+import pydantic
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+
+
+class ErrorMessage(UniversalBaseModel):
+    request_id: typing.Optional[str] = pydantic.Field(default=None)
+    """
+    The request ID associated with the error, if applicable.
+    """
+
+    message: str = pydantic.Field()
+    """
+    Human-readable error message describing what went wrong.
+    """
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow

cartesia/stt/types/flush_done_message.py

@@ -0,0 +1,26 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ...core.pydantic_utilities import UniversalBaseModel
+import pydantic
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+import typing
+
+
+class FlushDoneMessage(UniversalBaseModel):
+    """
+    Acknowledgment message sent in response to a `finalize` command, indicating that all buffered audio has been flushed and processed.
+    """
+
+    request_id: str = pydantic.Field()
+    """
+    Unique identifier for this transcription session.
+    """
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow

cartesia/stt/types/streaming_transcription_response.py

@@ -0,0 +1,94 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from __future__ import annotations
+from ...core.pydantic_utilities import UniversalBaseModel
+import typing
+from .transcription_word import TranscriptionWord
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+import pydantic
+
+
+class StreamingTranscriptionResponse_Transcript(UniversalBaseModel):
+    """
+    The server sends transcription results, control messages, or error messages. Each message has a `type` field to distinguish between different message types.
+    """
+
+    type: typing.Literal["transcript"] = "transcript"
+    request_id: str
+    text: str
+    is_final: bool
+    duration: typing.Optional[float] = None
+    language: typing.Optional[str] = None
+    words: typing.Optional[typing.List[TranscriptionWord]] = None
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow
+
+
+class StreamingTranscriptionResponse_FlushDone(UniversalBaseModel):
+    """
+    The server sends transcription results, control messages, or error messages. Each message has a `type` field to distinguish between different message types.
+    """
+
+    type: typing.Literal["flush_done"] = "flush_done"
+    request_id: str
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow
+
+
+class StreamingTranscriptionResponse_Done(UniversalBaseModel):
+    """
+    The server sends transcription results, control messages, or error messages. Each message has a `type` field to distinguish between different message types.
+    """
+
+    type: typing.Literal["done"] = "done"
+    request_id: str
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow
+
+
+class StreamingTranscriptionResponse_Error(UniversalBaseModel):
+    """
+    The server sends transcription results, control messages, or error messages. Each message has a `type` field to distinguish between different message types.
+    """
+
+    type: typing.Literal["error"] = "error"
+    request_id: typing.Optional[str] = None
+    message: str
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow
+
+
+StreamingTranscriptionResponse = typing.Union[
+    StreamingTranscriptionResponse_Transcript,
+    StreamingTranscriptionResponse_FlushDone,
+    StreamingTranscriptionResponse_Done,
+    StreamingTranscriptionResponse_Error,
+]

cartesia/stt/types/stt_encoding.py

@@ -0,0 +1,7 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+SttEncoding = typing.Union[
+    typing.Literal["pcm_s16le", "pcm_s32le", "pcm_f16le", "pcm_f32le", "pcm_mulaw", "pcm_alaw"], typing.Any
+]

cartesia/stt/types/timestamp_granularity.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+TimestampGranularity = typing.Union[typing.Literal["word"], typing.Any]

cartesia/stt/types/transcript_message.py

@@ -0,0 +1,50 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ...core.pydantic_utilities import UniversalBaseModel
+import pydantic
+import typing
+from .transcription_word import TranscriptionWord
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+
+
+class TranscriptMessage(UniversalBaseModel):
+    request_id: str = pydantic.Field()
+    """
+    Unique identifier for this transcription session.
+    """
+
+    text: str = pydantic.Field()
+    """
+    The transcribed text. May be partial or final depending on is_final.
+    
+    **Note**: Text may be empty in initial responses while the system accumulates sufficient audio for transcription. This is normal behavior - wait for responses with non-empty text or monitor is_final for completion status.
+    """
+
+    is_final: bool = pydantic.Field()
+    """
+    Whether this is a final transcription result or an interim result.
+    """
+
+    duration: typing.Optional[float] = pydantic.Field(default=None)
+    """
+    The duration of the audio transcribed so far, in seconds.
+    """
+
+    language: typing.Optional[str] = pydantic.Field(default=None)
+    """
+    The specified language of the input audio.
+    """
+
+    words: typing.Optional[typing.List[TranscriptionWord]] = pydantic.Field(default=None)
+    """
+    Word-level timestamps showing the start and end time of each word in seconds. Always included in streaming responses.
+    """
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow

cartesia/stt/types/transcription_response.py

@@ -0,0 +1,38 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ...core.pydantic_utilities import UniversalBaseModel
+import pydantic
+import typing
+from .transcription_word import TranscriptionWord
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+
+
+class TranscriptionResponse(UniversalBaseModel):
+    text: str = pydantic.Field()
+    """
+    The transcribed text.
+    """
+
+    language: typing.Optional[str] = pydantic.Field(default=None)
+    """
+    The specified language of the input audio.
+    """
+
+    duration: typing.Optional[float] = pydantic.Field(default=None)
+    """
+    The duration of the input audio in seconds.
+    """
+
+    words: typing.Optional[typing.List[TranscriptionWord]] = pydantic.Field(default=None)
+    """
+    Word-level timestamps showing the start and end time of each word. Only included when `[word]` is passed into `timestamp_granularities[]`.
+    """
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow

cartesia/stt/types/transcription_word.py

@@ -0,0 +1,32 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ...core.pydantic_utilities import UniversalBaseModel
+import pydantic
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+import typing
+
+
+class TranscriptionWord(UniversalBaseModel):
+    word: str = pydantic.Field()
+    """
+    The transcribed word.
+    """
+
+    start: float = pydantic.Field()
+    """
+    Start time of the word in seconds.
+    """
+
+    end: float = pydantic.Field()
+    """
+    End time of the word in seconds.
+    """
+
+    if IS_PYDANTIC_V2:
+        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
+    else:
+
+        class Config:
+            frozen = True
+            smart_union = True
+            extra = pydantic.Extra.allow

cartesia/tts/__init__.py

@@ -19,11 +19,13 @@ from .types import (
     RawEncoding,
     RawOutputFormat,
     Speed,
+    SseOutputFormat,
     SupportedLanguage,
     TtsRequest,
     TtsRequestEmbeddingSpecifier,
     TtsRequestIdSpecifier,
     TtsRequestVoiceSpecifier,
+    TtssseRequest,
     WavOutputFormat,
     WebSocketBaseResponse,
     WebSocketChunkResponse,
@@ -58,10 +60,12 @@ from .requests import (
     PhonemeTimestampsParams,
     RawOutputFormatParams,
     SpeedParams,
+    SseOutputFormatParams,
     TtsRequestEmbeddingSpecifierParams,
     TtsRequestIdSpecifierParams,
     TtsRequestParams,
     TtsRequestVoiceSpecifierParams,
+    TtssseRequestParams,
     WavOutputFormatParams,
     WebSocketBaseResponseParams,
     WebSocketChunkResponseParams,
@@ -115,6 +119,8 @@ __all__ = [
     "RawOutputFormatParams",
     "Speed",
     "SpeedParams",
+    "SseOutputFormat",
+    "SseOutputFormatParams",
     "SupportedLanguage",
     "TtsRequest",
     "TtsRequestEmbeddingSpecifier",
@@ -124,6 +130,8 @@ __all__ = [
     "TtsRequestParams",
     "TtsRequestVoiceSpecifier",
     "TtsRequestVoiceSpecifierParams",
+    "TtssseRequest",
+    "TtssseRequestParams",
     "WavOutputFormat",
     "WavOutputFormatParams",
     "WebSocketBaseResponse",

cartesia/tts/client.py

@@ -10,6 +10,8 @@ from ..core.request_options import RequestOptions
 from ..core.serialization import convert_and_respect_annotation_metadata
 from json.decoder import JSONDecodeError
 from ..core.api_error import ApiError
+from .requests.sse_output_format import SseOutputFormatParams
+from .types.context_id import ContextId
 from .types.web_socket_response import WebSocketResponse
 import httpx_sse
 from ..core.pydantic_utilities import parse_obj_as
@@ -119,10 +121,14 @@ class TtsClient:
         model_id: str,
         transcript: str,
         voice: TtsRequestVoiceSpecifierParams,
-        output_format: OutputFormatParams,
+        output_format: SseOutputFormatParams,
         language: typing.Optional[SupportedLanguage] = OMIT,
         duration: typing.Optional[float] = OMIT,
         speed: typing.Optional[ModelSpeed] = OMIT,
+        add_timestamps: typing.Optional[bool] = OMIT,
+        add_phoneme_timestamps: typing.Optional[bool] = OMIT,
+        use_normalized_timestamps: typing.Optional[bool] = OMIT,
+        context_id: typing.Optional[ContextId] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[WebSocketResponse]:
         """
@@ -135,7 +141,7 @@ class TtsClient:
 
         voice : TtsRequestVoiceSpecifierParams
 
-        output_format : OutputFormatParams
+        output_format : SseOutputFormatParams
 
         language : typing.Optional[SupportedLanguage]
 
@@ -145,6 +151,18 @@ class TtsClient:
 
         speed : typing.Optional[ModelSpeed]
 
+        add_timestamps : typing.Optional[bool]
+            Whether to return word-level timestamps. If `false` (default), no word timestamps will be produced at all. If `true`, the server will return timestamp events containing word-level timing information.
+
+        add_phoneme_timestamps : typing.Optional[bool]
+            Whether to return phoneme-level timestamps. If `false` (default), no phoneme timestamps will be produced - if `add_timestamps` is `true`, the produced timestamps will be word timestamps instead. If `true`, the server will return timestamp events containing phoneme-level timing information.
+
+        use_normalized_timestamps : typing.Optional[bool]
+            Whether to use normalized timestamps (True) or original timestamps (False).
+
+        context_id : typing.Optional[ContextId]
+            Optional context ID for this request.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -165,9 +183,9 @@ class TtsClient:
             voice={"mode": "id", "id": "694f9389-aac1-45b6-b726-9d9369183238"},
             language="en",
             output_format={
+                "container": "raw",
                 "sample_rate": 44100,
                 "encoding": "pcm_f32le",
-                "container": "raw",
             },
         )
         for chunk in response:
@@ -184,10 +202,14 @@ class TtsClient:
                 ),
                 "language": language,
                 "output_format": convert_and_respect_annotation_metadata(
-                    object_=output_format, annotation=OutputFormatParams, direction="write"
+                    object_=output_format, annotation=SseOutputFormatParams, direction="write"
                 ),
                 "duration": duration,
                 "speed": speed,
+                "add_timestamps": add_timestamps,
+                "add_phoneme_timestamps": add_phoneme_timestamps,
+                "use_normalized_timestamps": use_normalized_timestamps,
+                "context_id": context_id,
             },
             request_options=request_options,
             omit=OMIT,
@@ -321,10 +343,14 @@ class AsyncTtsClient:
         model_id: str,
         transcript: str,
         voice: TtsRequestVoiceSpecifierParams,
-        output_format: OutputFormatParams,
+        output_format: SseOutputFormatParams,
         language: typing.Optional[SupportedLanguage] = OMIT,
         duration: typing.Optional[float] = OMIT,
         speed: typing.Optional[ModelSpeed] = OMIT,
+        add_timestamps: typing.Optional[bool] = OMIT,
+        add_phoneme_timestamps: typing.Optional[bool] = OMIT,
+        use_normalized_timestamps: typing.Optional[bool] = OMIT,
+        context_id: typing.Optional[ContextId] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[WebSocketResponse]:
         """
@@ -337,7 +363,7 @@ class AsyncTtsClient:
 
         voice : TtsRequestVoiceSpecifierParams
 
-        output_format : OutputFormatParams
+        output_format : SseOutputFormatParams
 
         language : typing.Optional[SupportedLanguage]
 
@@ -347,6 +373,18 @@ class AsyncTtsClient:
 
         speed : typing.Optional[ModelSpeed]
 
+        add_timestamps : typing.Optional[bool]
+            Whether to return word-level timestamps. If `false` (default), no word timestamps will be produced at all. If `true`, the server will return timestamp events containing word-level timing information.
+
+        add_phoneme_timestamps : typing.Optional[bool]
+            Whether to return phoneme-level timestamps. If `false` (default), no phoneme timestamps will be produced - if `add_timestamps` is `true`, the produced timestamps will be word timestamps instead. If `true`, the server will return timestamp events containing phoneme-level timing information.
+
+        use_normalized_timestamps : typing.Optional[bool]
+            Whether to use normalized timestamps (True) or original timestamps (False).
+
+        context_id : typing.Optional[ContextId]
+            Optional context ID for this request.
+
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
@@ -372,9 +410,9 @@ class AsyncTtsClient:
                 voice={"mode": "id", "id": "694f9389-aac1-45b6-b726-9d9369183238"},
                 language="en",
                 output_format={
+                    "container": "raw",
                     "sample_rate": 44100,
                     "encoding": "pcm_f32le",
-                    "container": "raw",
                 },
             )
             async for chunk in response:
@@ -394,10 +432,14 @@ class AsyncTtsClient:
                 ),
                 "language": language,
                 "output_format": convert_and_respect_annotation_metadata(
-                    object_=output_format, annotation=OutputFormatParams, direction="write"
+                    object_=output_format, annotation=SseOutputFormatParams, direction="write"
                 ),
                 "duration": duration,
                 "speed": speed,
+                "add_timestamps": add_timestamps,
+                "add_phoneme_timestamps": add_phoneme_timestamps,
+                "use_normalized_timestamps": use_normalized_timestamps,
+                "context_id": context_id,
             },
             request_options=request_options,
             omit=OMIT,