livekit-plugins-vakyam 1.8.0__tar.gz

livekit_plugins_vakyam-1.8.0/.gitignore

@@ -0,0 +1,180 @@
+**/.vscode
+**/.DS_Store
+.env
+
+# 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
+
+# trunk
+.trunk/
+
+# 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
+docs/
+
+# Database files
+*.db
+
+
+# Examples for development
+examples/dev/*

livekit_plugins_vakyam-1.8.0/PKG-INFO

@@ -0,0 +1,80 @@
+Metadata-Version: 2.5
+Name: livekit-plugins-vakyam
+Version: 1.8.0
+Summary: LiveKit Agents plugin for Vakyam AI TTS (Raaga 1) — Indian-language text-to-speech
+Project-URL: Documentation, https://docs.livekit.io
+Project-URL: Website, https://livekit.io/
+Project-URL: Source, https://github.com/livekit/agents
+Author-email: LiveKit <hello@livekit.io>
+License-Expression: Apache-2.0
+Keywords: audio,indian-languages,livekit,raaga,realtime,text-to-speech,tts,vakyam,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 :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Topic :: Multimedia :: Sound/Audio
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10.0
+Requires-Dist: livekit-agents[codecs]>=1.8.0
+Requires-Dist: websockets<16.0,>=14.0
+Description-Content-Type: text/markdown
+
+# Vakyam AI plugin for LiveKit Agents
+
+Support for voice synthesis with [Vakyam AI](https://vakyam.ai/) Raaga 1 —
+text-to-speech for Indian languages.
+
+See [https://docs.vakyam.ai/integrations/livekit](https://docs.vakyam.ai/integrations/livekit)
+for provider docs.
+
+## Installation
+
+```bash
+pip install livekit-plugins-vakyam
+```
+
+Or with the LiveKit Agents extra:
+
+```bash
+uv add "livekit-agents[vakyam]"
+```
+
+## Pre-requisites
+
+You'll need an API key from [Vakyam](https://dashboard.vakyam.ai/api-keys).
+Set it as an environment variable:
+
+```bash
+export VAKYAM_API_KEY="vak_live_..."
+```
+
+## Usage
+
+```python
+from livekit.agents import AgentSession
+from livekit.plugins import vakyam
+
+session = AgentSession(
+    tts=vakyam.TTS(
+        model="raaga-v1",
+        voice="Archana",
+        language="ta-IN",
+        sample_rate=24000,
+    ),
+    # ... stt, llm, vad
+)
+```
+
+`stream()` uses the realtime WebSocket API and sentence-tokenizes LLM text so
+each utterance is one complete sentence (Vakyam does not accept partial
+tokens). `synthesize()` uses HTTP streaming (`POST /v1/tts/stream`) and
+returns PCM audio.
+
+WebSocket connections are pooled and reused between sequential agent turns.
+Each active synthesis stream has exclusive ownership of its connection, so an
+overlapping stream uses a separate connection. On interruption, the plugin
+sends `cancel`, drains through Vakyam's cancellation acknowledgement, and
+returns the healthy connection to the pool.

livekit_plugins_vakyam-1.8.0/README.md

@@ -0,0 +1,56 @@
+# Vakyam AI plugin for LiveKit Agents
+
+Support for voice synthesis with [Vakyam AI](https://vakyam.ai/) Raaga 1 —
+text-to-speech for Indian languages.
+
+See [https://docs.vakyam.ai/integrations/livekit](https://docs.vakyam.ai/integrations/livekit)
+for provider docs.
+
+## Installation
+
+```bash
+pip install livekit-plugins-vakyam
+```
+
+Or with the LiveKit Agents extra:
+
+```bash
+uv add "livekit-agents[vakyam]"
+```
+
+## Pre-requisites
+
+You'll need an API key from [Vakyam](https://dashboard.vakyam.ai/api-keys).
+Set it as an environment variable:
+
+```bash
+export VAKYAM_API_KEY="vak_live_..."
+```
+
+## Usage
+
+```python
+from livekit.agents import AgentSession
+from livekit.plugins import vakyam
+
+session = AgentSession(
+    tts=vakyam.TTS(
+        model="raaga-v1",
+        voice="Archana",
+        language="ta-IN",
+        sample_rate=24000,
+    ),
+    # ... stt, llm, vad
+)
+```
+
+`stream()` uses the realtime WebSocket API and sentence-tokenizes LLM text so
+each utterance is one complete sentence (Vakyam does not accept partial
+tokens). `synthesize()` uses HTTP streaming (`POST /v1/tts/stream`) and
+returns PCM audio.
+
+WebSocket connections are pooled and reused between sequential agent turns.
+Each active synthesis stream has exclusive ownership of its connection, so an
+overlapping stream uses a separate connection. On interruption, the plugin
+sends `cancel`, drains through Vakyam's cancellation acknowledgement, and
+returns the healthy connection to the pool.

livekit_plugins_vakyam-1.8.0/livekit/plugins/vakyam/__init__.py

@@ -0,0 +1,45 @@
+# 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.
+
+"""Vakyam AI plugin for LiveKit Agents
+
+Support for text-to-speech with [Vakyam AI](https://vakyam.ai/) Raaga 1.
+
+See https://docs.vakyam.ai/guides/realtime-websocket for protocol details.
+"""
+
+from .tts import TTS, ChunkedStream, SynthesizeStream
+from .version import __version__
+
+__all__ = ["TTS", "ChunkedStream", "SynthesizeStream", "__version__"]
+
+from livekit.agents import Plugin
+
+from .log import logger
+
+
+class VakyamPlugin(Plugin):
+    def __init__(self) -> None:
+        super().__init__(__name__, __version__, __package__, logger)
+
+
+Plugin.register_plugin(VakyamPlugin())
+
+_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_vakyam-1.8.0/livekit/plugins/vakyam/_utils.py

@@ -0,0 +1,214 @@
+# 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 json
+from typing import Any
+from urllib.parse import urlparse
+
+from livekit.agents import APIStatusError
+
+from .models import (
+    CUSTOM_VOICE_PREFIX,
+    MAX_SPEED,
+    MAX_TEXT_CHARACTERS,
+    MIN_SPEED,
+    SUPPORTED_LANGUAGES,
+    SUPPORTED_MODELS,
+    SUPPORTED_SAMPLE_RATES,
+    TTS_STREAM_PATH,
+    TTS_WEBSOCKET_PATH,
+)
+
+_RETRYABLE_WS_CODES = {
+    "rate_limit_exceeded",
+    "concurrency_limit_exceeded",
+    "tts_workers_busy",
+    "tts_workers_unconfigured",
+    "tts_workers_unavailable",
+    "internal_error",
+}
+
+
+def normalize_voice(voice: str) -> str:
+    """Normalize a preset voice name or custom ``vc_`` voice ID."""
+    if not isinstance(voice, str) or not voice.strip():
+        raise ValueError("voice must be a non-empty string")
+    normalized = voice.strip()
+    if normalized == CUSTOM_VOICE_PREFIX:
+        raise ValueError("custom voice IDs must include a value after 'vc_'")
+    return normalized
+
+
+def normalize_base_url(base_url: str, *, allow_insecure_base_url: bool = False) -> str:
+    """Normalize and validate a public API base URL."""
+    normalized = base_url.rstrip("/")
+    parsed = urlparse(normalized)
+    if parsed.scheme != "http":
+        return normalized
+
+    host = parsed.hostname or ""
+    if allow_insecure_base_url or host in {"localhost", "127.0.0.1", "::1"}:
+        return normalized
+
+    raise ValueError(
+        "base_url must use HTTPS unless it points to localhost. "
+        "Pass allow_insecure_base_url=True only for trusted development networks."
+    )
+
+
+def websocket_url(base_url: str) -> str:
+    """Build the Vakyam TTS WebSocket URL from an HTTP(S) or WS(S) base URL."""
+    root = base_url.rstrip("/")
+    if root.startswith("https://"):
+        root = "wss://" + root[len("https://") :]
+    elif root.startswith("http://"):
+        root = "ws://" + root[len("http://") :]
+    elif not root.startswith(("ws://", "wss://")):
+        root = "wss://" + root
+    if root.endswith(TTS_WEBSOCKET_PATH):
+        return root
+    return root + TTS_WEBSOCKET_PATH
+
+
+def http_stream_url(base_url: str) -> str:
+    """Build the Vakyam HTTP streaming TTS URL."""
+    return base_url.rstrip("/") + TTS_STREAM_PATH
+
+
+def validate_tts_options(
+    *,
+    model: str,
+    language: str,
+    sample_rate: int,
+    speed: float,
+    voice: str,
+) -> None:
+    """Validate constructor / update_options values."""
+    if model not in SUPPORTED_MODELS:
+        valid = ", ".join(sorted(SUPPORTED_MODELS))
+        raise ValueError(f"model '{model}' is not supported. Valid values are: {valid}.")
+    if language not in SUPPORTED_LANGUAGES:
+        valid = ", ".join(sorted(SUPPORTED_LANGUAGES))
+        raise ValueError(f"language '{language}' is not supported. Valid values are: {valid}.")
+    if sample_rate not in SUPPORTED_SAMPLE_RATES:
+        valid = ", ".join(str(v) for v in sorted(SUPPORTED_SAMPLE_RATES))
+        raise ValueError(
+            f"sample_rate '{sample_rate}' is not supported. Valid values are: {valid}."
+        )
+    if not MIN_SPEED <= speed <= MAX_SPEED:
+        raise ValueError(f"speed must be between {MIN_SPEED} and {MAX_SPEED}")
+    normalize_voice(voice)
+
+
+def validate_text(text: str) -> None:
+    """Validate utterance text for a single synthesis request."""
+    if not isinstance(text, str) or not text:
+        raise ValueError("text is required")
+    character_count = len(text)
+    if character_count > MAX_TEXT_CHARACTERS:
+        raise ValueError(
+            f"Input text is {character_count} characters. Maximum allowed is "
+            f"{MAX_TEXT_CHARACTERS} Unicode characters."
+        )
+
+
+def split_text(text: str, *, max_characters: int = MAX_TEXT_CHARACTERS) -> list[str]:
+    """Split an oversized utterance at whitespace, falling back to a hard boundary."""
+    if max_characters <= 0:
+        raise ValueError("max_characters must be greater than zero")
+
+    remaining = text.strip()
+    chunks: list[str] = []
+    while len(remaining) > max_characters:
+        split_at = remaining.rfind(" ", 0, max_characters + 1)
+        if split_at <= 0:
+            split_at = max_characters
+        chunk = remaining[:split_at].strip()
+        if chunk:
+            chunks.append(chunk)
+        remaining = remaining[split_at:].lstrip()
+    if remaining:
+        chunks.append(remaining)
+    return chunks
+
+
+def speech_payload(
+    *,
+    text: str,
+    model: str,
+    voice: str,
+    language: str,
+    sample_rate: int,
+    speed: float,
+    output_format: str = "pcm",
+) -> dict[str, Any]:
+    """JSON body for HTTP generate/stream requests."""
+    validate_text(text)
+    validate_tts_options(
+        model=model, language=language, sample_rate=sample_rate, speed=speed, voice=voice
+    )
+    return {
+        "text": text,
+        "model_id": model,
+        "voice": normalize_voice(voice),
+        "language": language,
+        "output_format": output_format,
+        "sample_rate": sample_rate,
+        "speed": speed,
+    }
+
+
+def raise_http_error(status: int, body: str) -> None:
+    """Raise ``APIStatusError`` from a Vakyam HTTP error envelope."""
+    parsed: object | None = None
+    try:
+        parsed = json.loads(body) if body else None
+    except json.JSONDecodeError:
+        parsed = None
+
+    error_code: str | None = None
+    if isinstance(parsed, dict) and isinstance(parsed.get("error"), dict):
+        error = parsed["error"]
+        code = error.get("code")
+        if isinstance(code, (str, int)):
+            error_code = str(code)
+
+    message = f"Vakyam TTS request failed with status {status}"
+    safe_body: dict[str, object] = {"status_code": status}
+    if error_code is not None:
+        message += f" (error code: {error_code})"
+        safe_body["error_code"] = error_code
+
+    raise APIStatusError(message, status_code=status, body=safe_body)
+
+
+def raise_ws_error(data: dict[str, Any]) -> None:
+    """Raise ``APIStatusError`` from a Vakyam WebSocket ``error`` frame."""
+    error = data.get("error") if isinstance(data.get("error"), dict) else {}
+    code = error.get("code") if isinstance(error, dict) else None
+    error_code = str(code) if isinstance(code, (str, int)) else None
+    retryable = error_code in _RETRYABLE_WS_CODES
+    message = "Vakyam TTS WebSocket request failed"
+    safe_body: dict[str, str] = {"type": "error"}
+    if error_code is not None:
+        message += f" (error code: {error_code})"
+        safe_body["code"] = error_code
+    raise APIStatusError(
+        message,
+        status_code=-1,
+        body=safe_body,
+        retryable=retryable,
+    )