(chore): update tests

src/elevenlabs/client.py

@@ -5,16 +5,14 @@
 from typing import Iterator, Optional, Union, \
   Optional, AsyncIterator
 
-from elevenlabs.types.model_response import ModelResponse
-
 from .base_client import \
   BaseElevenLabs, AsyncBaseElevenLabs
 from .core import RequestOptions, ApiError
-from .types import VoiceResponse, VoiceSettings, \
-  PronunciationDictionaryVersionLocator
+from .types import Voice, VoiceSettings, \
+  PronunciationDictionaryVersionLocator, Model
 
 
-DEFAULT_VOICE = VoiceResponse(
+DEFAULT_VOICE = Voice(
     voice_id="EXAVITQu4vr4xnSDxMaL",
     name="Rachel",
     settings=VoiceSettings(
@@ -66,9 +64,9 @@ def clone(
       name: str,
       files: typing.List[str],
       description: str,
-      labels: str,
+      labels: typing.Optional[str] = OMIT,
       request_options: typing.Optional[RequestOptions] = None
-    ) -> VoiceResponse:
+    ) -> Voice:
         """
           This is a manually maintained helper function that clones a voice from a set of audio files.
           **NOTE**: This function is a helper function and is simply making 
@@ -100,9 +98,9 @@ def generate(
       self,
       *,
       text: Union[str, Iterator[str]],
-      voice: Union[VoiceId, VoiceName, VoiceResponse] = DEFAULT_VOICE,
+      voice: Union[VoiceId, VoiceName, Voice] = DEFAULT_VOICE,
       voice_settings: typing.Optional[VoiceSettings] = DEFAULT_VOICE.settings,
-      model: Union[ModelId, ModelResponse] = "eleven_monolingual_v1",
+      model: Union[ModelId, Model] = "eleven_monolingual_v1",
       optimize_streaming_latency: typing.Optional[int] = 0,
       stream: bool = False,
       output_format: Optional[str] = "mp3_44100_128",
@@ -161,22 +159,23 @@ def generate(
             voice_id = next((v.voice_id for v in voices_response.voices if v.name == voice), None)
             if not voice_id:
                 raise ApiError(body=f"Voice {voice} not found.")
-        elif isinstance(voice, VoiceResponse):
+        elif isinstance(voice, Voice):
             voice_id = voice.voice_id
-            if voice_settings != DEFAULT_VOICE.settings:
+            if voice_settings != DEFAULT_VOICE.settings \
+                    and voice.settings is not None:
                 voice = voice.settings
         else: 
             voice_id = DEFAULT_VOICE.voice_id
 
         if isinstance(model, str):
             model_id = model
-        elif isinstance(model, ModelResponse):
+        elif isinstance(model, Model):
             model_id = model.model_id
 
 
-        if stream:
+        if not stream:
             if isinstance(text, str):
-                return self.text_to_speech.convert_as_stream(
+                return self.text_to_speech.convert(
                     voice_id=voice_id,
                     voice_settings=voice_settings,
                     optimize_streaming_latency=optimize_streaming_latency,
@@ -187,7 +186,7 @@ def generate(
                     model_id=model_id
                 )
             elif isinstance(text, Iterator):
-                return self.text_to_speech.convert_as_stream_input(
+                return self.text_to_speech.convert_as_stream(
                     voice_id=voice_id,
                     model_id=model_id,
                     voice_settings=voice_settings,
@@ -245,7 +244,7 @@ async def clone(
       description: str,
       labels: str,
       request_options: typing.Optional[RequestOptions] = None
-    ) -> VoiceResponse:
+    ) -> Voice:
         """
           This is a manually mnaintained helper function that generates a 
           voice from provided text.
@@ -280,9 +279,9 @@ async def generate(
       self,
       *,
       text: Union[str, Iterator[str]],
-      voice: Union[VoiceId, VoiceName, VoiceResponse] = DEFAULT_VOICE,
+      voice: Union[VoiceId, VoiceName, Voice] = DEFAULT_VOICE,
       voice_settings: typing.Optional[VoiceSettings] = DEFAULT_VOICE.settings,
-      model: Union[ModelId, ModelResponse] = "eleven_monolingual_v1",
+      model: Union[ModelId, Model] = "eleven_monolingual_v1",
       optimize_streaming_latency: typing.Optional[int] = 0,
       stream: bool = False,
       output_format: Optional[str] = "mp3_44100_128",
@@ -348,21 +347,22 @@ async def generate(
             voice_id = next((v.voice_id for v in voices_response.voices if v.name == voice), None)
             if not voice_id:
                 raise ApiError(body=f"Voice {voice} not found.")
-        elif isinstance(voice, VoiceResponse):
+        elif isinstance(voice, Voice):
             voice_id = voice.voice_id
-            if voice_settings != DEFAULT_VOICE.settings:
+            if voice_settings != DEFAULT_VOICE.settings \
+                    and voice.settings is not None:
                 voice = voice.settings
         else: 
             voice_id = DEFAULT_VOICE.voice_id
 
         if isinstance(model, str):
             model_id = model
-        elif isinstance(model, ModelResponse):
+        elif isinstance(model, Model):
             model_id = model.model_id
 
-        if stream:
+        if not stream:
             if isinstance(text, str):
-                return await self.text_to_speech.convert_as_stream(
+                return await self.text_to_speech.convert(
                     voice_id=voice_id,
                     model_id=model_id,
                     voice_settings=voice_settings,
@@ -373,7 +373,7 @@ async def generate(
                     pronunciation_dictionary_locators=pronunciation_dictionary_locators
                 )
             elif isinstance(text, Iterator):
-                return await self.text_to_speech.convert_as_stream_input(
+                return await self.text_to_speech.convert_as_stream(
                     voice_id=voice_id,
                     model_id=model_id,
                     voice_settings=voice_settings,

src/elevenlabs/play.py

@@ -1,6 +1,6 @@
 import shutil
 import subprocess
-from typing import Iterator
+from typing import Iterator, Union
 
 
 def is_installed(lib_name: str) -> bool:
@@ -10,7 +10,13 @@ def is_installed(lib_name: str) -> bool:
     return True
 
 
-def play(audio: bytes, notebook: bool = False, use_ffmpeg: bool = True) -> None:
+def play(
+    audio: Union[bytes, Iterator[bytes]], 
+    notebook: bool = False, 
+    use_ffmpeg: bool = True
+) -> None:
+    if isinstance(audio, Iterator):
+        audio = b"".join(audio)
     if notebook:
         from IPython.display import Audio, display  # type: ignore
 

tests/test_client.py

@@ -1,21 +1,24 @@
-import pytest
-
-from elevenlabs import generate, voices, Voice, VoiceSettings, play, stream
-from .utils import IN_GITHUB
+from elevenlabs import play, \
+    Voice, VoiceSettings, stream
+from .utils import IN_GITHUB, client
 
 
 def test_voices() -> None:
-    print("Voices are...", voices())
+    print("Voices are...", client.voices.get_all())
 
 
 def test_generate() -> None:
-    audio = generate(
+    audio = client.generate(
         text="Hello! My name is Bella.",
         voice=Voice(
             voice_id='EXAVITQu4vr4xnSDxMaL',
-            settings=VoiceSettings(stability=0.71, similarity_boost=0.5, style=0.0, use_speaker_boost=True)
-        )
-    )
+            settings=VoiceSettings(
+                stability=0.71,
+                similarity_boost=0.5,
+                style=0.0,
+                use_speaker_boost=True
+            )
+        ))
     if not IN_GITHUB:
         play(audio)  # type: ignore
 
@@ -25,27 +28,12 @@ def text_stream():
         yield "Hi there, I'm Eleven "
         yield "I'm a text to speech API "
 
-    audio_stream = generate(
+    audio_stream = client.generate(
         text=text_stream(),
         voice="Nicole",
         model="eleven_monolingual_v1",
         stream=True
     )
-
-    if not IN_GITHUB:
-        stream(audio_stream)  # type: ignore
-
-
-def test_generate_with_settings() -> None: 
-    from elevenlabs import Voice, VoiceSettings, generate
-
-    audio = generate(
-        text="Hello! My name is Bella.",
-        voice=Voice(
-            voice_id='EXAVITQu4vr4xnSDxMaL',
-            settings=VoiceSettings(stability=0.71, similarity_boost=0.5, style=0.0, use_speaker_boost=True)
-        )
-    )
 
     if not IN_GITHUB:
-        play(audio)  # type: ignore
+        stream(audio_stream)  # type: ignore

tests/test_history.py

@@ -1,40 +1,34 @@
-from pydoc import cli
-from .utils import IN_GITHUB
+import time
+from random import randint
 
+from elevenlabs import History, \
+    play
+
+from .utils import IN_GITHUB, client
 
-def test_history():
-    from elevenlabs import History
-    from elevenlabs.client import ElevenLabs
-
-    client = ElevenLabs()
 
+def test_history():
     page_size = 5
-    # Test that we can get history
     history = client.history.get_all(page_size=page_size)
     assert isinstance(history, History)
 
 
 def test_history_item_delete():
-    import time
-    from random import randint
-
-    from elevenlabs import generate, play
-    from elevenlabs.client import ElevenLabs
-
-    # Random text
     text = f"Test {randint(0, 1000)}"
-    audio = generate(text=text)  # Generate a history item to delete
+    audio = client.generate(text=text)
     if not IN_GITHUB:
-        play(audio)
+        play(audio)  # type: ignore
+
     time.sleep(1)
 
-    client = ElevenLabs()
     history = client.history.get_all().history
     print(history)
     history_item = history[0]
+
     # Check that item matches
     assert history_item.text == text
     client.history.delete(history_item.history_item_id)
+
     # Test that the history item was deleted
     history = client.history.get_all(page_size=1).history
-    assert len(history) == 0 or history[0].text != text
+    assert len(history) == 0 or history[0].text != text

tests/test_model.py

@@ -1,12 +1,8 @@
-import pytest
+from elevenlabs import Model
+from .utils import client
 
 
 def test_model():
-    from elevenlabs import Model
-    from elevenlabs.client import ElevenLabs
-
-    client = ElevenLabs()
-
     # Test that we can get all models
     models = client.models.get_all()
     print(models)

tests/test_voice.py

@@ -1,18 +1,15 @@
-from elevenlabs.client import ElevenLabs
-from elevenlabs.types.voice_response import VoiceResponse
-from .utils import IN_GITHUB
+from elevenlabs import Voice, \
+    VoiceSettings, play
+from .utils import IN_GITHUB, as_local_files, client
 
 
 def test_voice_from_id():
-    from elevenlabs import Voice, VoiceSettings
-    from elevenlabs.client import ElevenLabs
 
     # Test that we can get a voice from id
     voice_id = "21m00Tcm4TlvDq8ikWAM"
 
-    client = ElevenLabs()
     voice = client.voices.get(voice_id)
-    assert isinstance(voice, VoiceResponse)
+    assert isinstance(voice, Voice)
 
     assert voice.voice_id == voice_id
     assert voice.name == "Rachel"
@@ -22,18 +19,12 @@ def test_voice_from_id():
 
 
 def test_voice_clone():
-    from elevenlabs import Voice, clone, generate, play
-
-    from .utils import as_local_files
-
-    client = ElevenLabs()
-
     voice_file_urls = [
         "https://user-images.githubusercontent.com/12028621/235474694-584f7103-dab2-4c39-bb9a-8e5f00be85da.webm",
     ]
 
     with as_local_files(voice_file_urls) as files:
-        voice = clone(
+        voice = client.clone(
             name="Alex",
             description=(
                 "An old American male voice with a slight hoarseness in his throat."
@@ -48,7 +39,7 @@ def test_voice_clone():
     assert voice.category == "cloned"
     assert len(voice.samples) == len(voice_file_urls)
 
-    audio = generate(
+    audio = client.generate(
         text="Voice clone test successful.",
         voice=voice,
     )
@@ -61,40 +52,29 @@ def test_voice_clone():
 
 
 def test_voice_design():
-    from elevenlabs import Accent, Age, Gender, Voice, generate, play
-    from elevenlabs.client import ElevenLabs
-
-    client = ElevenLabs()
-
-    audio = client.voices.design(
-        name="Lexa",
+    audio = client.voice_generation.generate(
         text=(
             "Hi! My name is Lexa, I'm a voice design test. I should have a middle aged"
             " female voice with a british accent. "
         ),
-        voice_description="Middle aged female with british accent.",
-        gender=Gender.FEMALE,
-        age=Age.MIDDLE_AGED,
-        accent=Accent.BRITISH,
+        gender="female",
+        age="middle_aged",
+        accent="british",
         accent_strength=1.5,
     )
 
     assert isinstance(audio, bytes) and len(audio) > 0
-    
+
     if not IN_GITHUB:
         play(audio)
 
 
-
 def test_voices():
-    from elevenlabs import voices, Voice
-
     # Test that we can get voices from api
-    eleven_voices = voices()
+    eleven_voices = client.voices.get_all()
 
     assert len(eleven_voices) > 0
     assert isinstance(eleven_voices[0], Voice)
 
     for voice in eleven_voices:
         assert isinstance(voice, Voice)
-