cartesia 1.4.0__py3-none-any.whl → 2.0.0a2__py3-none-any.whl

cartesia/tts/types/web_socket_response.py

@@ -0,0 +1,124 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from __future__ import annotations
+from ...core.pydantic_utilities import UniversalBaseModel
+import typing
+from .context_id import ContextId
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+import pydantic
+from .flush_id import FlushId
+from .word_timestamps import WordTimestamps
+from .phoneme_timestamps import PhonemeTimestamps
+
+
+class WebSocketResponse_Chunk(UniversalBaseModel):
+    type: typing.Literal["chunk"] = "chunk"
+    data: str
+    step_time: float
+    context_id: typing.Optional[ContextId] = None
+    status_code: int
+    done: bool
+
+    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 WebSocketResponse_FlushDone(UniversalBaseModel):
+    type: typing.Literal["flush_done"] = "flush_done"
+    flush_id: FlushId
+    flush_done: bool
+    context_id: typing.Optional[ContextId] = None
+    status_code: int
+    done: bool
+
+    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 WebSocketResponse_Done(UniversalBaseModel):
+    type: typing.Literal["done"] = "done"
+    context_id: typing.Optional[ContextId] = None
+    status_code: int
+    done: bool
+
+    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 WebSocketResponse_Timestamps(UniversalBaseModel):
+    type: typing.Literal["timestamps"] = "timestamps"
+    word_timestamps: typing.Optional[WordTimestamps] = None
+    context_id: typing.Optional[ContextId] = None
+    status_code: int
+    done: bool
+
+    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 WebSocketResponse_Error(UniversalBaseModel):
+    type: typing.Literal["error"] = "error"
+    error: str
+    context_id: typing.Optional[ContextId] = None
+    status_code: int
+    done: bool
+
+    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 WebSocketResponse_PhonemeTimestamps(UniversalBaseModel):
+    type: typing.Literal["phoneme_timestamps"] = "phoneme_timestamps"
+    phoneme_timestamps: typing.Optional[PhonemeTimestamps] = None
+    context_id: typing.Optional[ContextId] = None
+    status_code: int
+    done: bool
+
+    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
+
+
+WebSocketResponse = typing.Union[
+    WebSocketResponse_Chunk,
+    WebSocketResponse_FlushDone,
+    WebSocketResponse_Done,
+    WebSocketResponse_Timestamps,
+    WebSocketResponse_Error,
+    WebSocketResponse_PhonemeTimestamps,
+]

cartesia/tts/types/web_socket_stream_options.py

@@ -0,0 +1,19 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ...core.pydantic_utilities import UniversalBaseModel
+import typing
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+import pydantic
+
+
+class WebSocketStreamOptions(UniversalBaseModel):
+    timeout: typing.Optional[float] = 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

cartesia/tts/types/web_socket_timestamps_response.py

@@ -0,0 +1,20 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from .web_socket_base_response import WebSocketBaseResponse
+import typing
+from .word_timestamps import WordTimestamps
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+import pydantic
+
+
+class WebSocketTimestampsResponse(WebSocketBaseResponse):
+    word_timestamps: typing.Optional[WordTimestamps] = 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

cartesia/tts/types/web_socket_tts_output.py

@@ -0,0 +1,27 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+import pydantic
+
+from ...core.pydantic_utilities import IS_PYDANTIC_V2, UniversalBaseModel
+from .context_id import ContextId
+from .flush_id import FlushId
+from .word_timestamps import WordTimestamps
+
+
+class WebSocketTtsOutput(UniversalBaseModel):
+    word_timestamps: typing.Optional[WordTimestamps] = None
+    audio: typing.Optional[bytes] = None
+    context_id: typing.Optional[ContextId] = None
+    flush_id: typing.Optional[FlushId] = None
+    flush_done: typing.Optional[bool] = 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

cartesia/tts/types/web_socket_tts_request.py

@@ -0,0 +1,36 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ...core.pydantic_utilities import UniversalBaseModel
+import pydantic
+import typing
+from .output_format import OutputFormat
+from .tts_request_voice_specifier import TtsRequestVoiceSpecifier
+import typing_extensions
+from ...core.serialization import FieldMetadata
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+
+
+class WebSocketTtsRequest(UniversalBaseModel):
+    model_id: str = pydantic.Field()
+    """
+    The ID of the model to use for the generation. See [Models](/build-with-sonic/models) for available models.
+    """
+
+    output_format: typing.Optional[OutputFormat] = None
+    transcript: typing.Optional[str] = None
+    voice: TtsRequestVoiceSpecifier
+    duration: typing.Optional[int] = None
+    language: typing.Optional[str] = None
+    add_timestamps: typing.Optional[bool] = None
+    add_phoneme_timestamps: typing.Optional[bool] = None
+    continue_: typing_extensions.Annotated[typing.Optional[bool], FieldMetadata(alias="continue")] = None
+    context_id: typing.Optional[str] = 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

cartesia/tts/types/word_timestamps.py

@@ -0,0 +1,21 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from ...core.pydantic_utilities import UniversalBaseModel
+import typing
+from ...core.pydantic_utilities import IS_PYDANTIC_V2
+import pydantic
+
+
+class WordTimestamps(UniversalBaseModel):
+    words: typing.List[str]
+    start: typing.List[float]
+    end: typing.List[float]
+
+    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/utils/tts.py

@@ -0,0 +1,64 @@
+import io
+import typing
+
+from pydub import AudioSegment  # type: ignore
+
+from .types import OutputFormatMapping
+
+
+def get_output_format(output_format_name: str):
+    """Convenience method to get the output_format dictionary from a given output format name.
+
+    Args:
+        output_format_name (str): The name of the output format.
+
+    Returns:
+        OutputFormat: A dictionary containing the details of the output format to be passed into tts.sse() or tts.websocket().send()
+
+    Raises:
+        ValueError: If the output_format name is not supported
+    """
+    if output_format_name in OutputFormatMapping._format_mapping:
+        output_format_obj = OutputFormatMapping.get_format(output_format_name)
+    else:
+        raise ValueError(f"Unsupported format: {output_format_name}")
+
+    return output_format_obj
+
+
+def concat_audio_segments(
+    left_audio: typing.Optional[bytes],
+    infill_audio: bytes,
+    right_audio: typing.Optional[bytes],
+    format: str = "wav",
+) -> bytes:
+    """Helper method to concatenate three audio segments while preserving audio format and headers.
+
+    Args:
+        left_audio: The audio segment that comes before the infill
+        infill_audio: The generated infill audio segment
+        right_audio: The audio segment that comes after the infill
+        format: The audio format (e.g., 'wav', 'mp3'). Defaults to 'wav'
+
+    Returns:
+        bytes: The concatenated audio as bytes
+
+    Raises:
+        ValueError: If the audio segments cannot be loaded or concatenated
+    """
+    try:
+        combined = AudioSegment.empty()
+        if left_audio:
+            combined += AudioSegment.from_file(io.BytesIO(left_audio), format=format)
+
+        combined += AudioSegment.from_file(io.BytesIO(infill_audio), format=format)
+
+        if right_audio:
+            combined += AudioSegment.from_file(io.BytesIO(right_audio), format=format)
+
+        output = io.BytesIO()
+        combined.export(output, format=format)
+        return output.getvalue()
+
+    except Exception as e:
+        raise ValueError(f"Failed to concatenate audio segments: {str(e)}")

cartesia/tts/utils/types.py

@@ -0,0 +1,70 @@
+class OutputFormatMapping:
+    _format_mapping = {
+        "raw_pcm_f32le_44100": {
+            "container": "raw",
+            "encoding": "pcm_f32le",
+            "sample_rate": 44100,
+        },
+        "raw_pcm_s16le_44100": {
+            "container": "raw",
+            "encoding": "pcm_s16le",
+            "sample_rate": 44100,
+        },
+        "raw_pcm_f32le_24000": {
+            "container": "raw",
+            "encoding": "pcm_f32le",
+            "sample_rate": 24000,
+        },
+        "raw_pcm_s16le_24000": {
+            "container": "raw",
+            "encoding": "pcm_s16le",
+            "sample_rate": 24000,
+        },
+        "raw_pcm_f32le_22050": {
+            "container": "raw",
+            "encoding": "pcm_f32le",
+            "sample_rate": 22050,
+        },
+        "raw_pcm_s16le_22050": {
+            "container": "raw",
+            "encoding": "pcm_s16le",
+            "sample_rate": 22050,
+        },
+        "raw_pcm_f32le_16000": {
+            "container": "raw",
+            "encoding": "pcm_f32le",
+            "sample_rate": 16000,
+        },
+        "raw_pcm_s16le_16000": {
+            "container": "raw",
+            "encoding": "pcm_s16le",
+            "sample_rate": 16000,
+        },
+        "raw_pcm_f32le_8000": {
+            "container": "raw",
+            "encoding": "pcm_f32le",
+            "sample_rate": 8000,
+        },
+        "raw_pcm_s16le_8000": {
+            "container": "raw",
+            "encoding": "pcm_s16le",
+            "sample_rate": 8000,
+        },
+        "raw_pcm_mulaw_8000": {
+            "container": "raw",
+            "encoding": "pcm_mulaw",
+            "sample_rate": 8000,
+        },
+        "raw_pcm_alaw_8000": {
+            "container": "raw",
+            "encoding": "pcm_alaw",
+            "sample_rate": 8000,
+        },
+    }
+
+    @classmethod
+    def get_format(cls, format_name):
+        if format_name in cls._format_mapping:
+            return cls._format_mapping[format_name]
+        else:
+            raise ValueError(f"Unsupported format: {format_name}")

cartesia/version.py

@@ -1 +1,3 @@
-__version__ = "1.4.0"
+from importlib import metadata
+
+__version__ = metadata.version("cartesia")

cartesia/voice_changer/__init__.py

@@ -0,0 +1,27 @@
+# This file was auto-generated by Fern from our API Definition.
+
+from .types import (
+    OutputFormatContainer,
+    StreamingResponse,
+    StreamingResponse_Chunk,
+    StreamingResponse_Done,
+    StreamingResponse_Error,
+)
+from .requests import (
+    StreamingResponseParams,
+    StreamingResponse_ChunkParams,
+    StreamingResponse_DoneParams,
+    StreamingResponse_ErrorParams,
+)
+
+__all__ = [
+    "OutputFormatContainer",
+    "StreamingResponse",
+    "StreamingResponseParams",
+    "StreamingResponse_Chunk",
+    "StreamingResponse_ChunkParams",
+    "StreamingResponse_Done",
+    "StreamingResponse_DoneParams",
+    "StreamingResponse_Error",
+    "StreamingResponse_ErrorParams",
+]