livekit-plugins-volcengine 1.0.0rc4__tar.gz

livekit_plugins_volcengine-1.0.0rc4/.gitignore

@@ -0,0 +1,179 @@
+# 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
+
+# UV
+#   Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control.
+#   This is especially recommended for binary packages to ensure reproducibility, and is more
+#   commonly ignored for libraries.
+#uv.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/latest/usage/project/#working-with-version-control
+.pdm.toml
+.pdm-python
+.pdm-build/
+
+# 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/
+
+# Ruff stuff:
+.ruff_cache/
+
+# PyPI configuration file
+.pypirc
+
+# tests
+test.py
+
+.python-version

livekit_plugins_volcengine-1.0.0rc4/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 王梦迪
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

livekit_plugins_volcengine-1.0.0rc4/PKG-INFO

@@ -0,0 +1,66 @@
+Metadata-Version: 2.4
+Name: livekit-plugins-volcengine
+Version: 1.0.0rc4
+Summary: LiveKit Agent Plugins for Volcengine
+Author-email: wangmengdi <790990241@qq.com>
+License-File: LICENSE
+Keywords: audio,livekit,realtime,video,webrtc
+Classifier: Intended Audience :: Developers
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+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
+Requires-Dist: livekit-agents~=1.0.0rc4
+Requires-Dist: pydantic>=2.0
+Requires-Dist: python-dotenv>=1.1.0
+Description-Content-Type: text/markdown
+
+# LiveKit Plugins Volcengine
+
+Agent Framework plugin for services from Volcengine(火山引擎). Currently supports [TTS](https://www.volcengine.com/docs/6561/79817)
+
+## Installation
+```python
+pip install livekit-plugins-volcengine
+```
+
+## Pre-requisites
+
+- Volcengine TTS environment variable: `VOLCENGINE_TTS_ACCESS_TOKEN`
+
+## Usage
+
+```python
+from livekit.agents import Agent, AgentSession, JobContext, cli, WorkerOptions
+from livekit.plugins import openai, volcengine, deepgram, silero
+from dotenv import load_dotenv
+
+
+async def entry_point(ctx: JobContext):
+    
+    await ctx.connect()
+    
+    agent = Agent(instructions="You are a helpful assistant.")
+
+    session = AgentSession(
+        vad=silero.VAD.load(),
+        stt=deepgram.STT(language="zh"),
+        ## app_id and cluster can be found in the Volcengine TTS console
+        tts=volcengine.TTS(app_id="xxx", cluster="xxx", streaming=True),
+        llm=openai.LLM(model="gpt-4o-mini"),
+    )
+    
+    await session.start(agent=agent, room=ctx.room)
+    
+    await session.generate_reply()
+
+if __name__ == "__main__":
+    load_dotenv()
+    cli.run_app(WorkerOptions(entrypoint_fnc=entry_point))
+```
+

livekit_plugins_volcengine-1.0.0rc4/README.md

@@ -0,0 +1,44 @@
+# LiveKit Plugins Volcengine
+
+Agent Framework plugin for services from Volcengine(火山引擎). Currently supports [TTS](https://www.volcengine.com/docs/6561/79817)
+
+## Installation
+```python
+pip install livekit-plugins-volcengine
+```
+
+## Pre-requisites
+
+- Volcengine TTS environment variable: `VOLCENGINE_TTS_ACCESS_TOKEN`
+
+## Usage
+
+```python
+from livekit.agents import Agent, AgentSession, JobContext, cli, WorkerOptions
+from livekit.plugins import openai, volcengine, deepgram, silero
+from dotenv import load_dotenv
+
+
+async def entry_point(ctx: JobContext):
+    
+    await ctx.connect()
+    
+    agent = Agent(instructions="You are a helpful assistant.")
+
+    session = AgentSession(
+        vad=silero.VAD.load(),
+        stt=deepgram.STT(language="zh"),
+        ## app_id and cluster can be found in the Volcengine TTS console
+        tts=volcengine.TTS(app_id="xxx", cluster="xxx", streaming=True),
+        llm=openai.LLM(model="gpt-4o-mini"),
+    )
+    
+    await session.start(agent=agent, room=ctx.room)
+    
+    await session.generate_reply()
+
+if __name__ == "__main__":
+    load_dotenv()
+    cli.run_app(WorkerOptions(entrypoint_fnc=entry_point))
+```
+

livekit_plugins_volcengine-1.0.0rc4/livekit/plugins/volcengine/__init__.py

@@ -0,0 +1,25 @@
+from .tts import TTS
+from .version import __version__
+
+__all__ = ["TTS", "__version__"]
+
+from livekit.agents import Plugin
+
+from .log import logger
+
+
+class VolcenginePlugin(Plugin):
+    def __init__(self):
+        super().__init__(__name__, __version__, __package__, logger)
+
+
+Plugin.register_plugin(VolcenginePlugin())
+
+# Cleanup docs of unexported modules
+_module = dir()
+NOT_IN_ALL = [m for m in _module if m not in __all__]
+
+__pdoc__ = {}
+
+for n in NOT_IN_ALL:
+    __pdoc__[n] = False

livekit_plugins_volcengine-1.0.0rc4/livekit/plugins/volcengine/log.py

@@ -0,0 +1,4 @@
+from logging import getLogger
+
+
+logger = getLogger("livekit.plugins.volcengine")

livekit_plugins_volcengine-1.0.0rc4/livekit/plugins/volcengine/tts.py

@@ -0,0 +1,460 @@
+from __future__ import annotations
+from typing import Dict, Literal, ByteString, List, Tuple
+import os
+import base64
+import gzip
+import json
+from dataclasses import dataclass
+
+from livekit.agents import (
+    tts,
+    utils,
+    APIConnectionError,
+    APITimeoutError,
+    APIStatusError,
+    APIConnectOptions,
+)
+from livekit.agents.types import DEFAULT_API_CONNECT_OPTIONS
+from pydantic import BaseModel, Field
+import aiohttp
+import asyncio
+
+
+class _TTSOptions(BaseModel):
+    app_id: str
+    cluster: str
+    access_token: str | None = None
+    voice_type: str = "BV001_V2_streaming"
+    base_url: str = "https://openspeech.bytedance.com/api/v1"
+    sample_rate: Literal[24000, 16000, 8000] = 24000
+    encoding: Literal["mp3", "pcm"] = "pcm"
+    speed: float = Field(1.0, ge=0.2, le=3.0)
+    volume: float = Field(1.0, gt=0.1, le=3.0)
+    pitch: float = Field(1.0, ge=0.1, le=3.0)
+
+    def get_http_url(self):
+        return f"{self.base_url}/tts"
+
+    def get_http_header(self):
+        if self.access_token is None:
+            self.access_token = os.getenv("VOLCENGINE_TTS_ACCESS_TOKEN")
+            if self.access_token is None:
+                raise ValueError("VOLCENGINE_TTS_ACCESS_TOKEN is not set")
+        return {
+            "Authorization": f"Bearer;{self.access_token}",
+        }
+
+    def get_http_query_params(self, text: str, uid: str | None = None) -> Dict:
+        if uid is None:
+            uid = utils.shortuuid()
+        request_json = {
+            "app": {
+                "appid": self.app_id,
+                "token": self.access_token,
+                "cluster": self.cluster,
+            },
+            "user": {"uid": uid},
+            "audio": {
+                "voice_type": self.voice_type,
+                "encoding": self.encoding,
+                "speed_ratio": self.speed,
+                "volume_ratio": self.volume,
+                "pitch_ratio": 1.0,
+                "rate": self.sample_rate,
+            },
+            "request": {
+                "reqid": utils.shortuuid(),
+                "text": text,
+                "text_type": "plain",
+                "operation": "query",
+                "with_frontend": self.pitch,
+                "frontend_type": "unitTson",
+            },
+        }
+        return request_json
+
+    def get_ws_url(self):
+        return f"{self.base_url}/tts/ws_binary"
+
+    def get_ws_query_params(self, text: str, uid: str | None = None) -> bytearray:
+        if uid is None:
+            uid = utils.shortuuid()
+        submit_request_json = {
+            "app": {
+                "appid": self.app_id,
+                "token": self.access_token,
+                "cluster": self.cluster,
+            },
+            "user": {"uid": uid},
+            "audio": {
+                "voice_type": self.voice_type,
+                "encoding": self.encoding,
+                "speed_ratio": self.speed,
+                "volume_ratio": self.volume,
+                "pitch_ratio": self.pitch,
+                "rate": self.sample_rate,
+            },
+            "request": {
+                "reqid": utils.shortuuid(),
+                "text": text,
+                "text_type": "plain",
+                "operation": "submit",
+                "with_frontend": 1,
+                "frontend_type": "unitTson",
+            },
+        }
+        default_header = bytearray(b"\x11\x10\x11\x00")
+        payload_bytes = str.encode(json.dumps(submit_request_json))
+        payload_bytes = gzip.compress(
+            payload_bytes
+        )  # if no compression, comment this line
+        full_client_request = bytearray(default_header)
+        full_client_request.extend(
+            (len(payload_bytes)).to_bytes(4, "big")
+        )  # payload size(4 bytes)
+        full_client_request.extend(payload_bytes)  # payload
+        return full_client_request
+
+    def get_ws_header(self):
+        if self.access_token is None:
+            self.access_token = os.getenv("VOLCENGINE_TTS_ACCESS_TOKEN")
+            if self.access_token is None:
+                raise ValueError("VOLCENGINE_TTS_ACCESS_TOKEN is not set")
+        return {
+            "Authorization": f"Bearer;{self.access_token}",
+        }
+
+
+class TTS(tts.TTS):
+    def __init__(
+        self,
+        app_id: str,
+        cluster: str,
+        access_token: str | None = None,
+        voice_type: str = "BV001_V2_streaming",
+        sample_rate: Literal[24000, 16000, 8000] = 24000,
+        streaming: bool = True,
+        http_session: aiohttp.ClientSession | None = None,
+        max_session_duration: float = 600,
+    ):
+        """VolcEngine TTS
+
+        Args:
+            app_id (str): the app id of the tts, you can get it from the console.
+            cluster (str): the cluster of the tts, you can get it from the console.
+            access_token (str | None, optional): the access token of the tts, if not provided, the value of the environment variable VOLCENGINE_TTS_ACCESS_TOKEN will be used. Defaults to None.
+            voice_type (str, optional): the voice type of the tts, you can get it from https://www.volcengine.com/docs/6561/97465. Defaults to "BV001_V2_streaming". if you want to use the streaming api, you must ensure the voice type is end with "_streaming".
+            sample_rate (Literal[24000, 16000, 8000], optional): the sample rate of the tts. Defaults to 24000.
+            streaming (bool, optional): whether to use the streaming api. Defaults to True.
+            http_session (aiohttp.ClientSession | None, optional): the http session to use. Defaults to None.
+            max_session_duration (float, optional): the max duration of the http session. Defaults to 600.
+        """
+        super().__init__(
+            capabilities=tts.TTSCapabilities(streaming=streaming),
+            sample_rate=sample_rate,
+            num_channels=1,
+        )
+        self._opts = _TTSOptions(
+            app_id=app_id,
+            cluster=cluster,
+            access_token=access_token,
+            voice_type=voice_type,
+            sample_rate=sample_rate,
+        )
+        self._session = http_session
+
+        self._pool = utils.ConnectionPool[aiohttp.ClientWebSocketResponse](
+            connect_cb=self._connect_ws,
+            close_cb=self._close_ws,
+            max_session_duration=max_session_duration,
+            mark_refreshed_on_get=True,
+        )
+
+    def _ensure_session(self) -> aiohttp.ClientSession:
+        if self._session is None:
+            self._session = utils.http_context.http_session()
+
+        return self._session
+
+    async def _connect_ws(self) -> aiohttp.ClientWebSocketResponse:
+        session = self._ensure_session()
+        url = self._opts.get_ws_url()
+        headers = self._opts.get_ws_header()
+        return await asyncio.wait_for(
+            session.ws_connect(url, headers=headers), self._conn_options.timeout
+        )
+
+    async def _close_ws(self, ws: aiohttp.ClientWebSocketResponse):
+        await ws.close()
+
+    def synthesize(
+        self, text, *, conn_options: APIConnectOptions = DEFAULT_API_CONNECT_OPTIONS
+    ) -> ChunedStream:
+        return ChunedStream(
+            opts=self._opts,
+            session=self._ensure_session(),
+            tts=self,
+            input_text=text,
+            conn_options=conn_options,
+        )
+
+    def stream(self, *, conn_options: APIConnectOptions = DEFAULT_API_CONNECT_OPTIONS):
+        return SynthesizeStream(
+            tts=self,
+            conn_options=conn_options,
+            opts=self._opts,
+            pool=self._pool,
+            session=self._ensure_session(),
+        )
+
+
+class ChunedStream(tts.ChunkedStream):
+    def __init__(
+        self,
+        *,
+        opts: _TTSOptions,
+        session: aiohttp.ClientSession,
+        tts: TTS,
+        input_text,
+        conn_options=None,
+    ):
+        super().__init__(tts=tts, input_text=input_text, conn_options=conn_options)
+        self._opts: _TTSOptions = opts
+        self._session = session
+
+    async def _run(self) -> None:
+        request_id = utils.shortuuid()
+        bstream = utils.audio.AudioByteStream(
+            sample_rate=self._opts.sample_rate, num_channels=1
+        )
+        data = self._opts.get_http_query_params(text=self._input_text)
+        headers = self._opts.get_http_header()
+        try:
+            async with self._session.post(
+                self._opts.get_http_url(),
+                json=data,
+                headers=headers,
+                timeout=aiohttp.ClientTimeout(
+                    total=30,
+                    sock_connect=self._conn_options.timeout,
+                ),
+            ) as resp:
+                resp.raise_for_status()
+                emitter = tts.SynthesizedAudioEmitter(
+                    event_ch=self._event_ch,
+                    request_id=request_id,
+                )
+                data = await resp.json()
+                if "data" in data:
+                    data = data["data"]
+                    data = base64.b64decode(data)
+                    frames = bstream.write(data)
+                    for frame in frames:
+                        emitter.push(frame)
+                    for frame in bstream.flush():
+                        emitter.push(frame)
+                    emitter.flush()
+        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=None,
+                body=None,
+            ) from e
+        except Exception as e:
+            raise APIConnectionError() from e
+        finally:
+            emitter.flush()
+
+
+class SynthesizeStream(tts.SynthesizeStream):
+    def __init__(
+        self,
+        *,
+        opts: _TTSOptions,
+        session: aiohttp.ClientSession,
+        pool: utils.ConnectionPool[aiohttp.ClientWebSocketResponse],
+        tts: TTS,
+        conn_options=None,
+    ):
+        super().__init__(tts=tts, conn_options=conn_options)
+        self._opts: _TTSOptions = opts
+        self._session = session
+        self._pool = pool
+
+    async def _run(self):
+        request_id = utils.shortuuid()
+
+        sentence_splitter = ChineseSentenceSplitter()
+        bstream = utils.audio.AudioByteStream(
+            sample_rate=self._opts.sample_rate,
+            num_channels=1,
+        )
+        emitter = tts.SynthesizedAudioEmitter(
+            event_ch=self._event_ch,
+            request_id=request_id,
+        )
+
+        async def _send_task(sentence: str, ws: aiohttp.ClientWebSocketResponse):
+            data = self._opts.get_ws_query_params(text=sentence)
+            await ws.send_bytes(data)
+
+        async def _recv_task(ws: aiohttp.ClientWebSocketResponse):
+            while True:
+                res = await ws.receive_bytes()
+                done, data = parse_response(res)
+                if data is not None:
+                    frames = bstream.write(data)
+                    for frame in frames:
+                        emitter.push(frame)
+                if done:
+                    for frame in bstream.flush():
+                        emitter.push(frame)
+                    emitter.flush()
+                    break
+
+        async for token in self._input_ch:
+            if isinstance(token, self._FlushSentinel):
+                sentences = sentence_splitter.process_text(text="", is_last=True)
+            else:
+                sentences = sentence_splitter.process_text(text=token, is_last=False)
+            for sentence in sentences:
+                if len(sentence) == 0:
+                    continue
+                async with self._pool.connection() as ws:
+                    await asyncio.gather(
+                        _send_task(sentence, ws),
+                        _recv_task(ws),
+                    )
+
+
+def parse_response(res) -> Tuple[bool, ByteString | None]:
+    header_size = res[0] & 0x0F
+    message_type = res[1] >> 4
+    message_type_specific_flags = res[1] & 0x0F
+    message_compression = res[2] & 0x0F
+    payload = res[header_size * 4 :]
+    if message_type == 0xB:  # audio-only server response
+        if message_type_specific_flags == 0:  # no sequence number as ACK
+            return False, None
+        else:
+            sequence_number = int.from_bytes(payload[:4], "big", signed=True)
+            payload = payload[8:]
+        if sequence_number < 0:
+            return True, payload
+        else:
+            return False, payload
+    elif message_type == 0xF:
+        error_msg = payload[8:]
+        if message_compression == 1:
+            error_msg = gzip.decompress(error_msg)
+        error_msg = str(error_msg, "utf-8")
+
+        return True, None
+    elif message_type == 0xC:
+        payload = payload[4:]
+        if message_compression == 1:
+            payload = gzip.decompress(payload)
+        return False, None
+    else:
+        return True, None
+
+
+@dataclass
+class ChineseSentenceSplitter:
+    buffer: str = ""
+    use_level2_threshold: int = 100
+    use_level3_threshold: int = 200
+
+    def process_text(
+        self,
+        text: str,
+        is_last: bool = False,
+        special_text: str | None = None,
+    ) -> List[str]:
+        self.buffer = self.buffer + text
+        if special_text is not None:
+            if self.buffer.endswith(special_text):
+                return [self.buffer]
+        sentences, indices = self.split_sentences(self.buffer)
+        assert len(sentences) == len(indices), (
+            "The number of sentences and indices do not match"
+        )
+        if not is_last:
+            if len(indices) != 0:
+                self.buffer = self.buffer[indices[-1] + 1 :]
+            return sentences
+        else:
+            if len(sentences) == 0:
+                sentences = [self.buffer]
+                self.buffer = ""
+                return sentences
+            if indices[-1] == len(self.buffer) - 1:
+                self.buffer = ""
+                return sentences
+            else:
+                self.buffer = ""
+                return sentences + [text[indices[-1] + 1 :]]
+
+    def split_sentences(self, text: str) -> List[str]:
+        indices = self.get_sentence_end_indices(text)
+        sentences = []
+        start = 0
+        for i in indices:
+            t = text[start : i + 1]
+            if len(t) > 0:
+                sentences.append(t)
+                start = i + 1
+        return sentences, indices
+
+    def is_sentence_end_level1(self, text: str) -> bool:
+        return text.endswith(
+            (
+                "!",
+                "?",
+                "。",
+                "？",
+                "！",
+                "；",
+                ";",
+            )
+        )
+
+    def is_sentence_end_level2(self, text: str) -> bool:
+        return text.endswith(
+            (
+                "、",
+                "...",
+                "…",
+                ",",
+                "，",
+            )
+        )
+
+    def is_sentence_end_level3(self, text: str) -> bool:
+        return text.endswith(
+            (
+                ":",
+                "：",
+            )
+        )
+
+    def get_sentence_end_indices(self, text: str) -> List[int]:
+        sents_l1 = [i for i, c in enumerate(text) if self.is_sentence_end_level1(c)]
+        if len(sents_l1) == 0 and len(text) > self.use_level2_threshold:
+            sents_l2 = [i for i, c in enumerate(text) if self.is_sentence_end_level2(c)]
+            if len(sents_l2) == 0 and len(text) > self.use_level3_threshold:
+                sents_l3 = [
+                    i for i, c in enumerate(text) if self.is_sentence_end_level3(c)
+                ]
+                return sents_l3
+            else:
+                return sents_l2
+
+        else:
+            return sents_l1
+
+    def reset(self):
+        self.buffer = ""

livekit_plugins_volcengine-1.0.0rc4/livekit/plugins/volcengine/version.py

@@ -0,0 +1 @@
+__version__ = "1.0.0.rc4"

livekit_plugins_volcengine-1.0.0rc4/pyproject.toml

@@ -0,0 +1,55 @@
+[project]
+name = "livekit-plugins-volcengine"
+dynamic = ["version"]
+description = "LiveKit Agent Plugins for Volcengine"
+readme = "README.md"
+authors = [
+    { name = "wangmengdi", email = "790990241@qq.com" }
+]
+keywords = ["webrtc", "realtime", "audio", "video", "livekit"]
+requires-python = ">=3.9"
+dependencies = [
+    "livekit-agents~=1.0.0rc4",
+    "pydantic>=2.0",
+    "python-dotenv>=1.1.0",
+]
+
+classifiers = [
+    "Intended Audience :: Developers",
+    "License :: OSI Approved :: Apache Software License",
+    "Topic :: Multimedia :: Sound/Audio",
+    "Topic :: Multimedia :: Video",
+    "Topic :: Scientific/Engineering :: Artificial Intelligence",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3 :: Only",
+]
+
+
+[dependency-groups]
+dev = [
+    "ipykernel>=6.29.5",
+    "livekit-plugins-cartesia>=1.0.0rc4",
+    "livekit-plugins-openai>=1.0.0rc4",
+    "livekit-plugins-deepgram>=1.0.0rc4",
+    "livekit-plugins-silero>=1.0.0rc4",
+]
+
+[[tool.uv.index]]
+name = "tsinghua"
+url = "https://pypi.tuna.tsinghua.edu.cn/simple"
+
+
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[tool.hatch.version]
+path = "livekit/plugins/volcengine/version.py"
+
+[tool.hatch.build.targets.wheel]
+packages = ["livekit"]
+
+[tool.hatch.build.targets.sdist]
+include = ["/livekit"]