livekit-plugins-resemble 0.1.0__tar.gz → 0.1.0rc1__tar.gz

livekit_plugins_resemble-0.1.0rc1/.gitignore

@@ -0,0 +1,168 @@
+**/.vscode
+**/.DS_Store
+
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+share/python-wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.nox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+*.py,cover
+.hypothesis/
+.pytest_cache/
+cover/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+db.sqlite3-journal
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+.pybuilder/
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# IPython
+profile_default/
+ipython_config.py
+
+# pyenv
+#   For a library or package, you might want to ignore these files since the code is
+#   intended to run in multiple environments; otherwise, check them in:
+# .python-version
+
+# pipenv
+#   According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
+#   However, in case of collaboration, if having platform-specific dependencies or dependencies
+#   having no cross-platform support, pipenv may install dependencies that don't work, or not
+#   install all needed dependencies.
+#Pipfile.lock
+
+# poetry
+#   Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#   https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control
+#poetry.lock
+
+# pdm
+#   Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control.
+#pdm.lock
+#   pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it
+#   in version control.
+#   https://pdm.fming.dev/#use-with-ide
+.pdm.toml
+
+# PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm
+__pypackages__/
+
+# Celery stuff
+celerybeat-schedule
+celerybeat.pid
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+.dmypy.json
+dmypy.json
+
+# Pyre type checker
+.pyre/
+
+# pytype static type analyzer
+.pytype/
+
+# Cython debug symbols
+cython_debug/
+
+# PyCharm
+#  JetBrains specific template is maintained in a separate JetBrains.gitignore that can
+#  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
+#  and can be added to the global gitignore or merged into this file.  For a more nuclear
+#  option (not recommended) you can uncomment the following to ignore the entire idea folder.
+.idea/
+
+node_modules
+
+credentials.json
+pyrightconfig.json

{livekit_plugins_resemble-0.1.0 → livekit_plugins_resemble-0.1.0rc1}/PKG-INFO

@@ -1,34 +1,25 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-resemble
-Version: 0.1.0
+Version: 0.1.0rc1
 Summary: LiveKit Agents Plugin for Resemble AI
-Home-page: https://github.com/livekit/agents
-License: Apache-2.0
 Project-URL: Documentation, https://docs.livekit.io
 Project-URL: Website, https://livekit.io/
 Project-URL: Source, https://github.com/livekit/agents
-Keywords: webrtc,realtime,audio,video,livekit,resemble,tts
+Author-email: LiveKit <hello@livekit.io>
+License-Expression: Apache-2.0
+Keywords: audio,livekit,realtime,video,webrtc
 Classifier: Intended Audience :: Developers
-Classifier: Topic :: Multimedia :: Sound/Audio
-Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Programming Language :: Python :: 3
-Classifier: Programming Language :: Python :: 3.11
-Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3 :: Only
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Topic :: Multimedia :: Sound/Audio
+Classifier: Topic :: Multimedia :: Video
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Requires-Python: >=3.9.0
+Requires-Dist: livekit-agents>=1.0.0.rc7
 Description-Content-Type: text/markdown
-Requires-Dist: livekit-agents[codecs]>=0.12.3
-Requires-Dist: websockets==12.0
-Dynamic: classifier
-Dynamic: description
-Dynamic: description-content-type
-Dynamic: home-page
-Dynamic: keywords
-Dynamic: license
-Dynamic: project-url
-Dynamic: requires-dist
-Dynamic: requires-python
-Dynamic: summary
 
 # LiveKit Plugins Resemble
 
@@ -147,4 +138,4 @@ This plugin uses two different approaches to generate speech:
 1. **One-off Synthesis** - Uses Resemble's REST API for simple text-to-speech conversion
 2. **Streaming Synthesis** - Uses Resemble's WebSocket API for real-time streaming synthesis
 
-The WebSocket streaming API is only available for Resemble AI Business plan users. 
+The WebSocket streaming API is only available for Resemble AI Business plan users. 

{livekit_plugins_resemble-0.1.0 → livekit_plugins_resemble-0.1.0rc1}/livekit/plugins/resemble/models.py

@@ -1,10 +1,5 @@
 from enum import Enum
 
 
-class OutputFormat(str, Enum):
-    WAV = "wav"
-    MP3 = "mp3"
-
-
 class Precision(str, Enum):
     PCM_16 = "PCM_16"

livekit_plugins_resemble-0.1.0rc1/livekit/plugins/resemble/tts.py

@@ -0,0 +1,452 @@
+# Copyright 2025 LiveKit, Inc.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+from __future__ import annotations
+
+import asyncio
+import base64
+import json
+import os
+import weakref
+from dataclasses import dataclass
+
+import aiohttp
+
+from livekit.agents import (
+    APIConnectionError,
+    APIConnectOptions,
+    APIStatusError,
+    APITimeoutError,
+    tokenize,
+    tts,
+    utils,
+)
+from livekit.agents.types import DEFAULT_API_CONNECT_OPTIONS
+
+from .log import logger
+
+RESEMBLE_WEBSOCKET_URL = "wss://websocket.cluster.resemble.ai/stream"
+RESEMBLE_REST_API_URL = "https://f.cluster.resemble.ai/synthesize"
+NUM_CHANNELS = 1
+DEFAULT_VOICE_UUID = "55592656"
+BUFFERED_WORDS_COUNT = 3
+
+
+@dataclass
+class _TTSOptions:
+    voice_uuid: str
+    sample_rate: int
+    tokenizer: tokenize.SentenceTokenizer
+
+
+class TTS(tts.TTS):
+    def __init__(
+        self,
+        *,
+        api_key: str | None = None,
+        voice_uuid: str | None = None,
+        tokenizer: tokenize.SentenceTokenizer | None = None,
+        sample_rate: int = 44100,
+        http_session: aiohttp.ClientSession | None = None,
+        use_streaming: bool = True,
+    ) -> None:
+        """
+        Create a new instance of the Resemble TTS.
+
+        See https://docs.app.resemble.ai/docs/text_to_speech/ for more documentation on all of these options.
+
+        Args:
+            voice_uuid (str, optional): The voice UUID for the desired voice. Defaults to None.
+            sample_rate (int, optional): The audio sample rate in Hz. Defaults to 44100.
+            api_key (str | None, optional): The Resemble API key. If not provided, it will be read from the RESEMBLE_API_KEY environment variable.
+            http_session (aiohttp.ClientSession | None, optional): An existing aiohttp ClientSession to use. If not provided, a new session will be created.
+            tokenizer (tokenize.SentenceTokenizer, optional): The tokenizer to use. Defaults to tokenize.SentenceTokenizer().
+            use_streaming (bool, optional): Whether to use streaming or not. Defaults to True.
+        """  # noqa: E501
+        super().__init__(
+            capabilities=tts.TTSCapabilities(streaming=use_streaming),
+            sample_rate=sample_rate,
+            num_channels=NUM_CHANNELS,
+        )
+
+        api_key = api_key or os.environ.get("RESEMBLE_API_KEY")
+        if not api_key:
+            raise ValueError(
+                "Resemble API key is required, either as argument or set RESEMBLE_API_KEY"
+                " environment variable"
+            )
+        self._api_key = api_key
+
+        if tokenizer is None:
+            tokenizer = tokenize.basic.SentenceTokenizer(min_sentence_len=BUFFERED_WORDS_COUNT)
+
+        if voice_uuid is None:
+            voice_uuid = DEFAULT_VOICE_UUID
+
+        self._opts = _TTSOptions(
+            voice_uuid=voice_uuid,
+            sample_rate=sample_rate,
+            tokenizer=tokenizer,
+        )
+
+        self._session = http_session
+        self._streams = weakref.WeakSet[SynthesizeStream]()
+        self._pool = utils.ConnectionPool[aiohttp.ClientWebSocketResponse](
+            connect_cb=self._connect_ws,
+            close_cb=self._close_ws,
+        )
+
+    async def _connect_ws(self) -> aiohttp.ClientWebSocketResponse:
+        session = self._ensure_session()
+
+        return await asyncio.wait_for(
+            session.ws_connect(
+                RESEMBLE_WEBSOCKET_URL,
+                headers={"Authorization": f"Bearer {self._api_key}"},
+            ),
+            self._conn_options.timeout,
+        )
+
+    async def _close_ws(self, ws: aiohttp.ClientWebSocketResponse):
+        await ws.close()
+
+    def _ensure_session(self) -> aiohttp.ClientSession:
+        if not self._session:
+            self._session = utils.http_context.http_session()
+
+        return self._session
+
+    def prewarm(self) -> None:
+        self._pool.prewarm()
+
+    def update_options(
+        self,
+        *,
+        voice_uuid: str | None = None,
+        sample_rate: int | None = None,
+    ) -> None:
+        """
+        Update the Text-to-Speech (TTS) configuration options.
+
+        Args:
+            voice_uuid (str, optional): The voice UUID for the desired voice.
+            sample_rate (int, optional): The audio sample rate in Hz.
+        """  # noqa: E501
+        self._opts.voice_uuid = voice_uuid or self._opts.voice_uuid
+        self._opts.sample_rate = sample_rate or self._opts.sample_rate
+
+    def synthesize(
+        self,
+        text: str,
+        *,
+        conn_options: APIConnectOptions | None = None,
+    ) -> ChunkedStream:
+        return ChunkedStream(
+            tts=self,
+            input_text=text,
+            conn_options=conn_options or DEFAULT_API_CONNECT_OPTIONS,
+            opts=self._opts,
+            api_key=self._api_key,
+            session=self._ensure_session(),
+        )
+
+    def stream(self, *, conn_options: APIConnectOptions | None = None) -> SynthesizeStream:
+        stream = SynthesizeStream(
+            tts=self,
+            pool=self._pool,
+            opts=self._opts,
+            api_key=self._api_key,
+        )
+        self._streams.add(stream)
+        return stream
+
+    async def aclose(self) -> None:
+        for stream in list(self._streams):
+            await stream.aclose()
+        self._streams.clear()
+        await self._pool.aclose()
+        await super().aclose()
+
+
+class ChunkedStream(tts.ChunkedStream):
+    """Synthesize text into speech in one go using Resemble AI's REST API."""
+
+    def __init__(
+        self,
+        *,
+        tts: TTS,
+        input_text: str,
+        opts: _TTSOptions,
+        conn_options: APIConnectOptions,
+        api_key: str,
+        session: aiohttp.ClientSession,
+    ) -> None:
+        super().__init__(tts=tts, input_text=input_text, conn_options=conn_options)
+        self._opts, self._session, self._api_key = opts, session, api_key
+
+    async def _run(self) -> None:
+        request_id = utils.shortuuid()
+
+        # Create request headers
+        headers = {
+            "Authorization": f"Bearer {self._api_key}",
+            "Content-Type": "application/json",
+            "Accept": "application/json",  # Expect JSON response
+        }
+
+        # Create request payload
+        payload = {
+            "voice_uuid": self._opts.voice_uuid,
+            "data": self._input_text,
+            "sample_rate": self._opts.sample_rate,
+            "precision": "PCM_16",
+        }
+        decoder = utils.codecs.AudioStreamDecoder(
+            sample_rate=self._opts.sample_rate,
+            num_channels=NUM_CHANNELS,
+        )
+
+        try:
+            async with self._session.post(
+                RESEMBLE_REST_API_URL,
+                headers=headers,
+                json=payload,
+                timeout=aiohttp.ClientTimeout(
+                    total=30,
+                    sock_connect=self._conn_options.timeout,
+                ),
+            ) as response:
+                response.raise_for_status()
+                response_json = await response.json()
+
+                # Check for success
+                if not response_json.get("success", False):
+                    issues = response_json.get("issues", ["Unknown error"])
+                    error_msg = "; ".join(issues)
+                    raise APIStatusError(
+                        message=f"Resemble API returned failure: {error_msg}",
+                        status_code=response.status,
+                        request_id=request_id,
+                        body=json.dumps(response_json),
+                    )
+
+                # Extract base64-encoded audio content
+                audio_content_b64 = response_json.get("audio_content")
+                if not audio_content_b64:
+                    raise APIStatusError(
+                        message="No audio content in response",
+                        status_code=response.status,
+                        request_id=request_id,
+                        body=json.dumps(response_json),
+                    )
+
+                # Decode base64 to get raw audio bytes
+                audio_bytes = base64.b64decode(audio_content_b64)
+                decoder.push(audio_bytes)
+                decoder.end_input()
+
+                emitter = tts.SynthesizedAudioEmitter(
+                    event_ch=self._event_ch,
+                    request_id=request_id,
+                )
+                async for frame in decoder:
+                    emitter.push(frame)
+                emitter.flush()
+
+        except aiohttp.ClientResponseError as e:
+            raise APIStatusError(
+                message=e.message,
+                status_code=e.status,
+                request_id=request_id,
+                body=f"resemble api error: {str(e)}",
+            ) from e
+        except asyncio.TimeoutError as e:
+            raise APITimeoutError() from e
+        except aiohttp.ClientError as e:
+            raise APIConnectionError(
+                message=f"Resemble API connection error: {str(e)}",
+            ) from e
+        except Exception as e:
+            raise APIConnectionError(f"Error during synthesis: {str(e)}") from e
+        finally:
+            await decoder.aclose()
+
+
+class SynthesizeStream(tts.SynthesizeStream):
+    """Stream-based text-to-speech synthesis using Resemble AI WebSocket API.
+
+
+    This implementation connects to Resemble's WebSocket API for real-time streaming
+    synthesis. Note that this requires a Business plan subscription with Resemble AI.
+    """
+
+    def __init__(
+        self,
+        *,
+        tts: TTS,
+        opts: _TTSOptions,
+        pool: utils.ConnectionPool[aiohttp.ClientWebSocketResponse],
+        api_key: str,
+    ):
+        super().__init__(tts=tts)
+        self._opts, self._pool, self._api_key = opts, pool, api_key
+
+    async def _run(self) -> None:
+        request_id = utils.shortuuid()
+        self._segments_ch = utils.aio.Chan[tokenize.SentenceStream]()
+
+        @utils.log_exceptions(logger=logger)
+        async def _tokenize_input():
+            """tokenize text from the input_ch to words"""
+            input_stream = None
+            async for input in self._input_ch:
+                if isinstance(input, str):
+                    if input_stream is None:
+                        # new segment (after flush for e.g)
+                        input_stream = self._opts.tokenizer.stream()
+                        self._segments_ch.send_nowait(input_stream)
+                    input_stream.push_text(input)
+                elif isinstance(input, self._FlushSentinel):
+                    if input_stream is not None:
+                        input_stream.end_input()
+                    input_stream = None
+            if input_stream is not None:
+                input_stream.end_input()
+            self._segments_ch.close()
+
+        @utils.log_exceptions(logger=logger)
+        async def _process_segments():
+            async for input_stream in self._segments_ch:
+                await self._run_ws(input_stream)
+
+        tasks = [
+            asyncio.create_task(_tokenize_input()),
+            asyncio.create_task(_process_segments()),
+        ]
+        try:
+            await asyncio.gather(*tasks)
+        except asyncio.TimeoutError as e:
+            raise APITimeoutError() from e
+        except aiohttp.ClientResponseError as e:
+            raise APIStatusError(
+                message=e.message,
+                status_code=e.status,
+                request_id=request_id,
+                body=None,
+            ) from e
+        except Exception as e:
+            raise APIConnectionError() from e
+        finally:
+            await utils.aio.gracefully_cancel(*tasks)
+
+    async def _run_ws(
+        self,
+        input_stream: tokenize.SentenceStream,
+    ) -> None:
+        async with self._pool.connection() as ws:
+            segment_id = utils.shortuuid()
+            decoder = utils.codecs.AudioStreamDecoder(
+                sample_rate=self._opts.sample_rate,
+                num_channels=NUM_CHANNELS,
+            )
+            index_lock = asyncio.Lock()
+            current_index = 0
+            pending_requests = set()
+
+            @utils.log_exceptions(logger=logger)
+            async def _send_task(ws: aiohttp.ClientWebSocketResponse):
+                nonlocal current_index
+                index = 0
+                async for data in input_stream:
+                    payload = {
+                        "voice_uuid": self._opts.voice_uuid,
+                        "data": data.token,
+                        "request_id": index,
+                        "sample_rate": self._opts.sample_rate,
+                        "precision": "PCM_16",
+                        "output_format": "mp3",
+                    }
+                    async with index_lock:
+                        pending_requests.add(index)
+                    index += 1
+                    current_index = index
+                    await ws.send_str(json.dumps(payload))
+
+            @utils.log_exceptions(logger=logger)
+            async def _emit_task():
+                emitter = tts.SynthesizedAudioEmitter(
+                    event_ch=self._event_ch,
+                    request_id=str(current_index),
+                    segment_id=segment_id,
+                )
+                async for frame in decoder:
+                    emitter.push(frame)
+                emitter.flush()
+
+            @utils.log_exceptions(logger=logger)
+            async def _recv_task(ws: aiohttp.ClientWebSocketResponse):
+                while True:
+                    msg = await ws.receive()
+                    if msg.type in (
+                        aiohttp.WSMsgType.CLOSED,
+                        aiohttp.WSMsgType.CLOSE,
+                        aiohttp.WSMsgType.CLOSING,
+                    ):
+                        raise APIStatusError(
+                            "Resemble connection closed unexpectedly",
+                            request_id=str(current_index),
+                        )
+
+                    if msg.type != aiohttp.WSMsgType.TEXT:
+                        logger.warning("Unexpected Resemble message type %s", msg.type)
+                        continue
+
+                    data = json.loads(msg.data)
+
+                    if data.get("type") == "audio":
+                        if data.get("audio_content", None):
+                            b64data = base64.b64decode(data["audio_content"])
+                            decoder.push(b64data)
+
+                    elif data.get("type") == "audio_end":
+                        async with index_lock:
+                            index = data["request_id"]
+                            pending_requests.remove(index)
+                            if not pending_requests:
+                                decoder.end_input()
+                                break  # we are not going to receive any more audio
+                    else:
+                        logger.error("Unexpected Resemble message %s", data)
+
+            tasks = [
+                asyncio.create_task(_send_task(ws)),
+                asyncio.create_task(_recv_task(ws)),
+                asyncio.create_task(_emit_task()),
+            ]
+
+            try:
+                await asyncio.gather(*tasks)
+            except asyncio.TimeoutError as e:
+                raise APITimeoutError() from e
+            except aiohttp.ClientResponseError as e:
+                raise APIStatusError(
+                    message=e.message,
+                    status_code=e.status,
+                    request_id=str(current_index),
+                    body=None,
+                ) from e
+            except Exception as e:
+                raise APIConnectionError() from e
+            finally:
+                await utils.aio.gracefully_cancel(*tasks)

{livekit_plugins_resemble-0.1.0 → livekit_plugins_resemble-0.1.0rc1}/livekit/plugins/resemble/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "0.1.0"
+__version__ = '0.1.0.rc1'