Implementation for livekit-plugins-neuphonic (TTS provider) (#1612)

Co-authored-by: Jayesh Parmar <60539217+jayeshp19@users.noreply.github.com>

Author: adnansiddiquei

README.md

@@ -102,6 +102,7 @@ We've partnered with OpenAI on a new `MultimodalAgent` API in the Agents framewo
 | Deepgram     | livekit-plugins-deepgram   | ✅        |               | [deepgram.TTS()](https://docs.livekit.io/python/livekit/plugins/deepgram/index.html#livekit.plugins.deepgram.TTS)       |
 | Play.ai      | livekit-plugins-playai     | ✅        | ✅            | [playai.TTS()](https://docs.livekit.io/python/livekit/plugins/playai/index.html#livekit.plugins.playai.TTS)             |
 | Rime         | livekit-plugins-rime       | ✅        |               | [rime.TTS()](https://docs.livekit.io/python/livekit/plugins/rime/index.html#livekit.plugins.rime.TTS)                   |
+| Neuphonic    | livekit-plugins-neuphonic  | ✅        | ✅            | neuphonic.TTS()                                                                                                       |
 | AWS Polly    | livekit-plugins-aws        | ✅        |               | [aws.TTS()](https://docs.livekit.io/python/livekit/plugins/aws/index.html#livekit.plugins.aws.TTS)                      |
 
 ### Other plugins

examples/text-to-speech/neuphonic_tts.py

@@ -0,0 +1,60 @@
+import asyncio
+import logging
+
+from dotenv import load_dotenv
+from livekit import rtc
+from livekit.agents import AutoSubscribe, JobContext, WorkerOptions, cli
+from livekit.plugins import neuphonic
+
+load_dotenv()
+
+logger = logging.getLogger("neuphonic-tts-demo")
+logger.setLevel(logging.INFO)
+
+
+async def entrypoint(job: JobContext):
+    logger.info("starting tts example agent")
+
+    SAMPLE_RATE = 22050
+    NUM_CHANNELS = 1
+
+    tts = neuphonic.TTS(
+        # voice_id=<uuid>,
+        sample_rate=SAMPLE_RATE  # defaults to 22050
+    )
+
+    source = rtc.AudioSource(SAMPLE_RATE, NUM_CHANNELS)
+    track = rtc.LocalAudioTrack.create_audio_track("agent-mic", source)
+    options = rtc.TrackPublishOptions()
+    options.source = rtc.TrackSource.SOURCE_MICROPHONE
+
+    await job.connect(auto_subscribe=AutoSubscribe.SUBSCRIBE_NONE)
+    publication = await job.room.local_participant.publish_track(track, options)
+    await publication.wait_for_subscription()
+
+    stream = tts.stream()
+
+    async def _playback_task():
+        async for audio in stream:
+            await source.capture_frame(audio.frame)
+
+    task = asyncio.create_task(_playback_task())
+
+    text = "Hello from Neuphonic. You have just successfully run the example!"
+
+    # split into two word chunks to simulate LLM streaming
+    words = text.split()
+    for i in range(0, len(words), 2):
+        chunk = " ".join(words[i : i + 2])
+        if chunk:
+            logger.info(f'pushing chunk: "{chunk} "')
+            stream.push_text(chunk + " ")
+
+    # Mark end of input segment
+    stream.flush()
+    stream.end_input()
+    await asyncio.gather(task)
+
+
+if __name__ == "__main__":
+    cli.run_app(WorkerOptions(entrypoint_fnc=entrypoint))

livekit-plugins/install_local.sh

@@ -22,4 +22,5 @@ pip install \
   "${SCRIPT_DIR}/livekit-plugins-turn-detector" \
   "${SCRIPT_DIR}/livekit-plugins-rime" \
   "${SCRIPT_DIR}/livekit-plugins-aws" \
-  "${SCRIPT_DIR}/livekit-plugins-speechmatics"
+  "${SCRIPT_DIR}/livekit-plugins-speechmatics" \
+  "${SCRIPT_DIR}/livekit-plugins-neuphonic"

livekit-plugins/install_plugins_editable.sh

@@ -23,5 +23,6 @@ pip install -e ./livekit-plugins-llama-index --config-settings editable_mode=str
 pip install -e ./livekit-plugins-turn-detector --config-settings editable_mode=strict
 pip install -e ./livekit-plugins-silero --config-settings editable_mode=strict
 pip install -e ./livekit-plugins-speechmatics --config-settings editable_mode=strict
+pip install -e ./livekit-plugins-neuphonic --config-settings editable_mode=strict
 pip install -e ./livekit-plugins-browser --config-settings editable_mode=strict
 

livekit-plugins/livekit-plugins-neuphonic/CHANGELOG.md

@@ -0,0 +1 @@
+# livekit-plugins-neuphonic

livekit-plugins/livekit-plugins-neuphonic/README.md

@@ -0,0 +1,13 @@
+# LiveKit Plugins Neuphonic
+
+Agent Framework plugin for voice synthesis with [Neuphonic](https://neuphonic.com) API.
+
+## Installation
+
+```bash
+pip install livekit-plugins-neuphonic
+```
+
+## Pre-requisites
+
+You'll need an API key from Neuphonic. It can be set as an environment variable: `NEUPHONIC_API_TOKEN`

livekit-plugins/livekit-plugins-neuphonic/livekit/plugins/neuphonic/__init__.py

@@ -0,0 +1,39 @@
+# Copyright 2023 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 .tts import TTS, ChunkedStream
+from .version import __version__
+
+__all__ = ["TTS", "ChunkedStream", "__version__"]
+
+from livekit.agents import Plugin
+
+from .log import logger
+
+
+class NeuphonicPlugin(Plugin):
+    def __init__(self):
+        super().__init__(__name__, __version__, __package__, logger)
+
+
+Plugin.register_plugin(NeuphonicPlugin())
+
+# 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/livekit-plugins-neuphonic/livekit/plugins/neuphonic/log.py

@@ -0,0 +1,3 @@
+import logging
+
+logger = logging.getLogger("livekit.plugins.neuphonic")

livekit-plugins/livekit-plugins-neuphonic/livekit/plugins/neuphonic/models.py

@@ -0,0 +1,10 @@
+from typing import Literal
+
+TTSEncodings = Literal[
+    "pcm_linear",
+    "pcm_mulaw",
+]
+
+TTSModels = Literal["neu-fast", "neu-hq"]
+
+TTSLangCodes = Literal["en", "nl", "es", "de", "hi", "en-hi", "ar"]