cartesia 2.0.0b1__tar.gz → 2.0.0b7__tar.gz

{cartesia-2.0.0b1 → cartesia-2.0.0b7}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cartesia
-Version: 2.0.0b1
+Version: 2.0.0b7
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers
@@ -57,7 +57,7 @@ A full reference for this library is available [here](./reference.md).
 from cartesia import Cartesia
 import os
 
-client = Cartesia(api_key=os.environ.get("CARTESIA_API_KEY"))
+client = Cartesia(api_key=os.getenv("CARTESIA_API_KEY"))
 
 # Get all available voices
 voices = client.voices.list()
@@ -65,21 +65,32 @@ print(voices)
 
 # Get a specific voice
 voice = client.voices.get(id="a0e99841-438c-4a64-b679-ae501e7d6091")
-print("The embedding for", voice["name"], "is", voice["embedding"])
+print("The embedding for", voice.name, "is", voice.embedding)
 
-# Clone a voice using filepath
-cloned_voice_embedding = client.voices.clone(filepath="path/to/voice")
+# Clone a voice using file data
+cloned_voice = client.voices.clone(
+    clip=open("path/to/voice.wav", "rb"),
+    name="Test cloned voice",
+    language="en",
+    mode="similarity",  # or "stability"
+    enhance=False, # use enhance=True to clean and denoise the cloning audio
+    description="Test voice description"
+)
 
 # Mix voices together
-mixed_voice_embedding = client.voices.mix(
-    [{ "id": "voice_id_1", "weight": 0.5 }, { "id": "voice_id_2", "weight": 0.25 }, { "id": "voice_id_3", "weight": 0.25 }]
+mixed_voice = client.voices.mix(
+    voices=[
+        {"id": "voice_id_1", "weight": 0.25},
+        {"id": "voice_id_2", "weight": 0.75}
+    ]
 )
 
-# Create a new voice
+# Create a new voice from embedding
 new_voice = client.voices.create(
-    name="New Voice",
-    description="A clone of my own voice",
-    embedding=cloned_voice_embedding,
+    name="Test Voice",
+    description="Test voice description",
+    embedding=[...],  # List[float] with 192 dimensions
+    language="en"
 )
 ```
 
@@ -90,15 +101,22 @@ Instantiate and use the client with the following:
 ```python
 from cartesia import Cartesia
 from cartesia.tts import OutputFormat_Raw, TtsRequestIdSpecifier
+import os
 
 client = Cartesia(
-    api_key="YOUR_API_KEY",
+    api_key=os.getenv("CARTESIA_API_KEY"),
 )
 client.tts.bytes(
-    model_id="sonic-english",
+    model_id="sonic-2",
     transcript="Hello, world!",
-    voice={"id": "694f9389-aac1-45b6-b726-9d9369183238"},
-    ),
+    voice={
+        "mode": "id",
+        "id": "694f9389-aac1-45b6-b726-9d9369183238",
+        "experimental_controls": {
+            "speed": 0.5,  # range between [-1.0, 1.0], or "slow", "fastest", etc.
+            "emotion": ["positivity", "curiosity:low"] # list of emotions with optional intensity
+        }
+    },
     language="en",
     output_format={
         "container": "raw",
@@ -114,18 +132,18 @@ The SDK also exports an `async` client so that you can make non-blocking calls t
 
 ```python
 import asyncio
+import os
 
 from cartesia import AsyncCartesia
 from cartesia.tts import OutputFormat_Raw, TtsRequestIdSpecifier
 
 client = AsyncCartesia(
-    api_key="YOUR_API_KEY",
+    api_key=os.getenv("CARTESIA_API_KEY"),
 )
 
-
 async def main() -> None:
-    await client.tts.bytes(
-        model_id="sonic-english",
+    async for output in client.tts.bytes(
+        model_id="sonic-2",
         transcript="Hello, world!",
         voice={"id": "694f9389-aac1-45b6-b726-9d9369183238"},
         language="en",
@@ -134,7 +152,8 @@ async def main() -> None:
             "sample_rate": 44100,
             "encoding": "pcm_f32le",
         },
-    )
+    ):
+        print(f"Received chunk of size: {len(output)}")
 
 
 asyncio.run(main())
@@ -162,26 +181,38 @@ The SDK supports streaming responses, as well, the response will be a generator
 ```python
 from cartesia import Cartesia
 from cartesia.tts import Controls, OutputFormat_RawParams, TtsRequestIdSpecifierParams
+import os
 
-client = Cartesia(
-    api_key="YOUR_API_KEY",
-)
-response = client.tts.sse(
-    model_id="string",
-    transcript="string",
-    voice={
-        "id": "string",
-        "experimental_controls": {
-            speed=1.1,
-            emotion="anger:lowest",
+def get_tts_chunks():
+    client = Cartesia(
+        api_key=os.getenv("CARTESIA_API_KEY"),
+    )
+    response = client.tts.sse(
+        model_id="sonic-2",
+        transcript="Hello world!",
+        voice={
+            "id": "f9836c6e-a0bd-460e-9d3c-f7299fa60f94",
+            "experimental_controls": {
+                "speed": "normal",
+                "emotion": [],
+            },
         },
-    },
-    language="en",
-    output_format={},
-    duration=1.1,
-)
-for chunk in response:
-    yield chunk
+        language="en",
+        output_format={
+            "container": "raw",
+            "encoding": "pcm_f32le",
+            "sample_rate": 44100,
+        },
+    )
+
+    audio_chunks = []
+    for chunk in response:
+        audio_chunks.append(chunk)
+    return audio_chunks
+
+chunks = get_tts_chunks()
+for chunk in chunks:
+    print(f"Received chunk of size: {len(chunk.data)}")
 ```
 
 ## WebSocket
@@ -190,16 +221,16 @@ for chunk in response:
 from cartesia import Cartesia
 from cartesia.tts import TtsRequestEmbeddingSpecifierParams, OutputFormat_RawParams
 import pyaudio
+import os
 
 client = Cartesia(
-    api_key="YOUR_API_KEY",
+    api_key=os.getenv("CARTESIA_API_KEY"),
 )
 voice_id = "a0e99841-438c-4a64-b679-ae501e7d6091"
-voice = client.voices.get(id=voice_id)
 transcript = "Hello! Welcome to Cartesia"
 
 # You can check out our models at https://docs.cartesia.ai/getting-started/available-models
-model_id = "sonic-english"
+model_id = "sonic-2"
 
 p = pyaudio.PyAudio()
 rate = 22050
@@ -213,11 +244,11 @@ ws = client.tts.websocket()
 for output in ws.send(
     model_id=model_id,
     transcript=transcript,
-    voice={"embedding": voice.embedding},
+    voice={"id": voice_id},
     stream=True,
     output_format={
         "container": "raw",
-        "encoding": "pcm_f32le", 
+        "encoding": "pcm_f32le",
         "sample_rate": 22050
     },
 ):
@@ -236,6 +267,55 @@ p.terminate()
 ws.close()  # Close the websocket connection
 ```
 
+## Requesting Timestamps
+
+```python
+import asyncio
+from cartesia import AsyncCartesia
+import os
+
+async def main():
+    client = AsyncCartesia(api_key=os.getenv("CARTESIA_API_KEY"))
+
+    # Connect to the websocket
+    ws = await client.tts.websocket()
+
+    # Generate speech with timestamps
+    output_generate = await ws.send(
+        model_id="sonic-2",
+        transcript="Hello! Welcome to Cartesia's text-to-speech.",
+        voice={"id": "f9836c6e-a0bd-460e-9d3c-f7299fa60f94"},
+        output_format={
+            "container": "raw",
+            "encoding": "pcm_f32le",
+            "sample_rate": 44100
+        },
+        add_timestamps=True,  # Enable word-level timestamps
+        stream=True
+    )
+
+    # Process the streaming response with timestamps
+    all_words = []
+    all_starts = []
+    all_ends = []
+    audio_chunks = []
+
+    async for out in output_generate:
+        # Collect audio data
+        if out.audio is not None:
+            audio_chunks.append(out.audio)
+
+        # Process timestamp data
+        if out.word_timestamps is not None:
+            all_words.extend(out.word_timestamps.words)    # List of words
+            all_starts.extend(out.word_timestamps.start)   # Start time for each word (seconds)
+            all_ends.extend(out.word_timestamps.end)       # End time for each word (seconds)
+
+    await ws.close()
+
+asyncio.run(main())
+```
+
 ## Advanced
 
 ### Retries
@@ -297,11 +377,33 @@ client = Cartesia(
 
 ## Contributing
 
-While we value open-source contributions to this SDK, this library is generated programmatically.
-Additions made directly to this library would have to be moved over to our generation code,
-otherwise they would be overwritten upon the next generated release. Feel free to open a PR as
-a proof of concept, but know that we will not be able to merge it as-is. We suggest opening
-an issue first to discuss with us!
+Note that most of this library is generated programmatically from
+<https://github.com/cartesia-ai/docs> — before making edits to a file, verify it's not autogenerated
+by checking for this comment at the top of the file:
+
+```
+# This file was auto-generated by Fern from our API Definition.
+```
+
+### Running tests
+
+```sh
+uv pip install -r requirements.txt
+uv run pytest -rP -vv tests/custom/test_client.py::test_get_voices
+```
+### Manually generating SDK code from docs
+
+Assuming all your repos are cloned into your home directory:
+
+```sh
+$ cd ~/docs
+$ fern generate --group python-sdk --log-level debug --api version-2024-11-13 --preview
+$ cd ~/cartesia-python
+$ git pull ~/docs/fern/apis/version-2024-11-13/.preview/fern-python-sdk
+$ git commit --amend -m "manually regenerate from docs" # optional
+```
+
+### Automatically generating new SDK releases
 
-On the other hand, contributions to the README are always very welcome!
+From https://github.com/cartesia-ai/docs click `Actions` then `Release Python SDK`. (Requires permissions.)
 

{cartesia-2.0.0b1 → cartesia-2.0.0b7}/README.md

@@ -25,7 +25,7 @@ A full reference for this library is available [here](./reference.md).
 from cartesia import Cartesia
 import os
 
-client = Cartesia(api_key=os.environ.get("CARTESIA_API_KEY"))
+client = Cartesia(api_key=os.getenv("CARTESIA_API_KEY"))
 
 # Get all available voices
 voices = client.voices.list()
@@ -33,21 +33,32 @@ print(voices)
 
 # Get a specific voice
 voice = client.voices.get(id="a0e99841-438c-4a64-b679-ae501e7d6091")
-print("The embedding for", voice["name"], "is", voice["embedding"])
+print("The embedding for", voice.name, "is", voice.embedding)
 
-# Clone a voice using filepath
-cloned_voice_embedding = client.voices.clone(filepath="path/to/voice")
+# Clone a voice using file data
+cloned_voice = client.voices.clone(
+    clip=open("path/to/voice.wav", "rb"),
+    name="Test cloned voice",
+    language="en",
+    mode="similarity",  # or "stability"
+    enhance=False, # use enhance=True to clean and denoise the cloning audio
+    description="Test voice description"
+)
 
 # Mix voices together
-mixed_voice_embedding = client.voices.mix(
-    [{ "id": "voice_id_1", "weight": 0.5 }, { "id": "voice_id_2", "weight": 0.25 }, { "id": "voice_id_3", "weight": 0.25 }]
+mixed_voice = client.voices.mix(
+    voices=[
+        {"id": "voice_id_1", "weight": 0.25},
+        {"id": "voice_id_2", "weight": 0.75}
+    ]
 )
 
-# Create a new voice
+# Create a new voice from embedding
 new_voice = client.voices.create(
-    name="New Voice",
-    description="A clone of my own voice",
-    embedding=cloned_voice_embedding,
+    name="Test Voice",
+    description="Test voice description",
+    embedding=[...],  # List[float] with 192 dimensions
+    language="en"
 )
 ```
 
@@ -58,15 +69,22 @@ Instantiate and use the client with the following:
 ```python
 from cartesia import Cartesia
 from cartesia.tts import OutputFormat_Raw, TtsRequestIdSpecifier
+import os
 
 client = Cartesia(
-    api_key="YOUR_API_KEY",
+    api_key=os.getenv("CARTESIA_API_KEY"),
 )
 client.tts.bytes(
-    model_id="sonic-english",
+    model_id="sonic-2",
     transcript="Hello, world!",
-    voice={"id": "694f9389-aac1-45b6-b726-9d9369183238"},
-    ),
+    voice={
+        "mode": "id",
+        "id": "694f9389-aac1-45b6-b726-9d9369183238",
+        "experimental_controls": {
+            "speed": 0.5,  # range between [-1.0, 1.0], or "slow", "fastest", etc.
+            "emotion": ["positivity", "curiosity:low"] # list of emotions with optional intensity
+        }
+    },
     language="en",
     output_format={
         "container": "raw",
@@ -82,18 +100,18 @@ The SDK also exports an `async` client so that you can make non-blocking calls t
 
 ```python
 import asyncio
+import os
 
 from cartesia import AsyncCartesia
 from cartesia.tts import OutputFormat_Raw, TtsRequestIdSpecifier
 
 client = AsyncCartesia(
-    api_key="YOUR_API_KEY",
+    api_key=os.getenv("CARTESIA_API_KEY"),
 )
 
-
 async def main() -> None:
-    await client.tts.bytes(
-        model_id="sonic-english",
+    async for output in client.tts.bytes(
+        model_id="sonic-2",
         transcript="Hello, world!",
         voice={"id": "694f9389-aac1-45b6-b726-9d9369183238"},
         language="en",
@@ -102,7 +120,8 @@ async def main() -> None:
             "sample_rate": 44100,
             "encoding": "pcm_f32le",
         },
-    )
+    ):
+        print(f"Received chunk of size: {len(output)}")
 
 
 asyncio.run(main())
@@ -130,26 +149,38 @@ The SDK supports streaming responses, as well, the response will be a generator
 ```python
 from cartesia import Cartesia
 from cartesia.tts import Controls, OutputFormat_RawParams, TtsRequestIdSpecifierParams
+import os
 
-client = Cartesia(
-    api_key="YOUR_API_KEY",
-)
-response = client.tts.sse(
-    model_id="string",
-    transcript="string",
-    voice={
-        "id": "string",
-        "experimental_controls": {
-            speed=1.1,
-            emotion="anger:lowest",
+def get_tts_chunks():
+    client = Cartesia(
+        api_key=os.getenv("CARTESIA_API_KEY"),
+    )
+    response = client.tts.sse(
+        model_id="sonic-2",
+        transcript="Hello world!",
+        voice={
+            "id": "f9836c6e-a0bd-460e-9d3c-f7299fa60f94",
+            "experimental_controls": {
+                "speed": "normal",
+                "emotion": [],
+            },
         },
-    },
-    language="en",
-    output_format={},
-    duration=1.1,
-)
-for chunk in response:
-    yield chunk
+        language="en",
+        output_format={
+            "container": "raw",
+            "encoding": "pcm_f32le",
+            "sample_rate": 44100,
+        },
+    )
+
+    audio_chunks = []
+    for chunk in response:
+        audio_chunks.append(chunk)
+    return audio_chunks
+
+chunks = get_tts_chunks()
+for chunk in chunks:
+    print(f"Received chunk of size: {len(chunk.data)}")
 ```
 
 ## WebSocket
@@ -158,16 +189,16 @@ for chunk in response:
 from cartesia import Cartesia
 from cartesia.tts import TtsRequestEmbeddingSpecifierParams, OutputFormat_RawParams
 import pyaudio
+import os
 
 client = Cartesia(
-    api_key="YOUR_API_KEY",
+    api_key=os.getenv("CARTESIA_API_KEY"),
 )
 voice_id = "a0e99841-438c-4a64-b679-ae501e7d6091"
-voice = client.voices.get(id=voice_id)
 transcript = "Hello! Welcome to Cartesia"
 
 # You can check out our models at https://docs.cartesia.ai/getting-started/available-models
-model_id = "sonic-english"
+model_id = "sonic-2"
 
 p = pyaudio.PyAudio()
 rate = 22050
@@ -181,11 +212,11 @@ ws = client.tts.websocket()
 for output in ws.send(
     model_id=model_id,
     transcript=transcript,
-    voice={"embedding": voice.embedding},
+    voice={"id": voice_id},
     stream=True,
     output_format={
         "container": "raw",
-        "encoding": "pcm_f32le", 
+        "encoding": "pcm_f32le",
         "sample_rate": 22050
     },
 ):
@@ -204,6 +235,55 @@ p.terminate()
 ws.close()  # Close the websocket connection
 ```
 
+## Requesting Timestamps
+
+```python
+import asyncio
+from cartesia import AsyncCartesia
+import os
+
+async def main():
+    client = AsyncCartesia(api_key=os.getenv("CARTESIA_API_KEY"))
+
+    # Connect to the websocket
+    ws = await client.tts.websocket()
+
+    # Generate speech with timestamps
+    output_generate = await ws.send(
+        model_id="sonic-2",
+        transcript="Hello! Welcome to Cartesia's text-to-speech.",
+        voice={"id": "f9836c6e-a0bd-460e-9d3c-f7299fa60f94"},
+        output_format={
+            "container": "raw",
+            "encoding": "pcm_f32le",
+            "sample_rate": 44100
+        },
+        add_timestamps=True,  # Enable word-level timestamps
+        stream=True
+    )
+
+    # Process the streaming response with timestamps
+    all_words = []
+    all_starts = []
+    all_ends = []
+    audio_chunks = []
+
+    async for out in output_generate:
+        # Collect audio data
+        if out.audio is not None:
+            audio_chunks.append(out.audio)
+
+        # Process timestamp data
+        if out.word_timestamps is not None:
+            all_words.extend(out.word_timestamps.words)    # List of words
+            all_starts.extend(out.word_timestamps.start)   # Start time for each word (seconds)
+            all_ends.extend(out.word_timestamps.end)       # End time for each word (seconds)
+
+    await ws.close()
+
+asyncio.run(main())
+```
+
 ## Advanced
 
 ### Retries
@@ -265,10 +345,32 @@ client = Cartesia(
 
 ## Contributing
 
-While we value open-source contributions to this SDK, this library is generated programmatically.
-Additions made directly to this library would have to be moved over to our generation code,
-otherwise they would be overwritten upon the next generated release. Feel free to open a PR as
-a proof of concept, but know that we will not be able to merge it as-is. We suggest opening
-an issue first to discuss with us!
+Note that most of this library is generated programmatically from
+<https://github.com/cartesia-ai/docs> — before making edits to a file, verify it's not autogenerated
+by checking for this comment at the top of the file:
+
+```
+# This file was auto-generated by Fern from our API Definition.
+```
+
+### Running tests
+
+```sh
+uv pip install -r requirements.txt
+uv run pytest -rP -vv tests/custom/test_client.py::test_get_voices
+```
+### Manually generating SDK code from docs
+
+Assuming all your repos are cloned into your home directory:
+
+```sh
+$ cd ~/docs
+$ fern generate --group python-sdk --log-level debug --api version-2024-11-13 --preview
+$ cd ~/cartesia-python
+$ git pull ~/docs/fern/apis/version-2024-11-13/.preview/fern-python-sdk
+$ git commit --amend -m "manually regenerate from docs" # optional
+```
+
+### Automatically generating new SDK releases
 
-On the other hand, contributions to the README are always very welcome!
+From https://github.com/cartesia-ai/docs click `Actions` then `Release Python SDK`. (Requires permissions.)

{cartesia-2.0.0b1 → cartesia-2.0.0b7}/pyproject.toml

@@ -3,7 +3,7 @@ name = "cartesia"
 
 [tool.poetry]
 name = "cartesia"
-version = "2.0.0b1"
+version = "2.0.0b7"
 description = ""
 readme = "README.md"
 authors = []

{cartesia-2.0.0b1 → cartesia-2.0.0b7}/src/cartesia/__init__.py

@@ -121,13 +121,14 @@ from .voices import (
     EmbeddingSpecifier,
     EmbeddingSpecifierParams,
     Gender,
+    GenderPresentation,
+    GetVoicesResponse,
+    GetVoicesResponseParams,
     IdSpecifier,
     IdSpecifierParams,
     LocalizeDialect,
     LocalizeDialectParams,
     LocalizeEnglishDialect,
-    LocalizePortugueseDialect,
-    LocalizeSpanishDialect,
     LocalizeTargetLanguage,
     LocalizeVoiceRequest,
     LocalizeVoiceRequestParams,
@@ -138,6 +139,7 @@ from .voices import (
     UpdateVoiceRequest,
     UpdateVoiceRequestParams,
     Voice,
+    VoiceExpandOptions,
     VoiceId,
     VoiceMetadata,
     VoiceMetadataParams,
@@ -175,15 +177,16 @@ __all__ = [
     "FilePurpose",
     "FlushId",
     "Gender",
+    "GenderPresentation",
     "GenerationRequest",
     "GenerationRequestParams",
+    "GetVoicesResponse",
+    "GetVoicesResponseParams",
     "IdSpecifier",
     "IdSpecifierParams",
     "LocalizeDialect",
     "LocalizeDialectParams",
     "LocalizeEnglishDialect",
-    "LocalizePortugueseDialect",
-    "LocalizeSpanishDialect",
     "LocalizeTargetLanguage",
     "LocalizeVoiceRequest",
     "LocalizeVoiceRequestParams",
@@ -235,6 +238,7 @@ __all__ = [
     "UpdateVoiceRequest",
     "UpdateVoiceRequestParams",
     "Voice",
+    "VoiceExpandOptions",
     "VoiceId",
     "VoiceMetadata",
     "VoiceMetadataParams",

{cartesia-2.0.0b1 → cartesia-2.0.0b7}/src/cartesia/base_client.py

@@ -5,14 +5,12 @@ from .environment import CartesiaEnvironment
 import httpx
 from .core.client_wrapper import SyncClientWrapper
 from .api_status.client import ApiStatusClient
-from .datasets.client import DatasetsClient
 from .infill.client import InfillClient
 from .tts.client import TtsClient
 from .voice_changer.client import VoiceChangerClient
 from .voices.client import VoicesClient
 from .core.client_wrapper import AsyncClientWrapper
 from .api_status.client import AsyncApiStatusClient
-from .datasets.client import AsyncDatasetsClient
 from .infill.client import AsyncInfillClient
 from .tts.client import AsyncTtsClient
 from .voice_changer.client import AsyncVoiceChangerClient
@@ -78,7 +76,6 @@ class BaseCartesia:
             timeout=_defaulted_timeout,
         )
         self.api_status = ApiStatusClient(client_wrapper=self._client_wrapper)
-        self.datasets = DatasetsClient(client_wrapper=self._client_wrapper)
         self.infill = InfillClient(client_wrapper=self._client_wrapper)
         self.tts = TtsClient(client_wrapper=self._client_wrapper)
         self.voice_changer = VoiceChangerClient(client_wrapper=self._client_wrapper)
@@ -144,7 +141,6 @@ class AsyncBaseCartesia:
             timeout=_defaulted_timeout,
         )
         self.api_status = AsyncApiStatusClient(client_wrapper=self._client_wrapper)
-        self.datasets = AsyncDatasetsClient(client_wrapper=self._client_wrapper)
         self.infill = AsyncInfillClient(client_wrapper=self._client_wrapper)
         self.tts = AsyncTtsClient(client_wrapper=self._client_wrapper)
         self.voice_changer = AsyncVoiceChangerClient(client_wrapper=self._client_wrapper)