add nvidia riva - fastpitch

examples/foundational/01c-fastpitch.py

@@ -0,0 +1,56 @@
+#
+# Copyright (c) 2024, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+import asyncio
+import aiohttp
+import os
+import sys
+
+from pipecat.frames.frames import EndFrame, TTSSpeakFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.task import PipelineTask
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.services.riva import FastpitchTTSService
+from pipecat.transports.services.daily import DailyParams, DailyTransport
+
+from runner import configure
+
+from loguru import logger
+
+from dotenv import load_dotenv
+
+load_dotenv(override=True)
+
+logger.remove(0)
+logger.add(sys.stderr, level="DEBUG")
+
+
+async def main():
+    async with aiohttp.ClientSession() as session:
+        (room_url, _) = await configure(session)
+
+        transport = DailyTransport(
+            room_url, None, "Say One Thing", DailyParams(audio_out_enabled=True)
+        )
+
+        tts = FastpitchTTSService(api_key=os.getenv("NVIDIA_API_KEY"))
+
+        runner = PipelineRunner()
+
+        task = PipelineTask(Pipeline([tts, transport.output()]))
+
+        # Register an event handler so we can play the audio when the
+        # participant joins.
+        @transport.event_handler("on_first_participant_joined")
+        async def on_first_participant_joined(transport, participant):
+            participant_name = participant.get("info", {}).get("userName", "")
+            await task.queue_frames([TTSSpeakFrame(f"Aloha, {participant_name}!"), EndFrame()])
+
+        await runner.run(task)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

examples/foundational/07r-interruptible-riva.py

@@ -0,0 +1,94 @@
+#
+# Copyright (c) 2024, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+import asyncio
+import os
+import sys
+
+import aiohttp
+from dotenv import load_dotenv
+from loguru import logger
+from runner import configure
+
+from pipecat.audio.vad.silero import SileroVADAnalyzer
+from pipecat.frames.frames import LLMMessagesFrame
+from pipecat.pipeline.pipeline import Pipeline
+from pipecat.pipeline.runner import PipelineRunner
+from pipecat.pipeline.task import PipelineParams, PipelineTask
+from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContext
+
+from pipecat.services.deepgram import DeepgramSTTService
+from pipecat.services.riva import FastpitchTTSService, ParakeetSTTService
+
+from pipecat.services.openai import OpenAILLMService
+from pipecat.transports.services.daily import DailyParams, DailyTransport
+
+load_dotenv(override=True)
+
+logger.remove(0)
+logger.add(sys.stderr, level="DEBUG")
+
+
+async def main():
+    async with aiohttp.ClientSession() as session:
+        (room_url, _) = await configure(session)
+
+        transport = DailyTransport(
+            room_url,
+            None,
+            "Respond bot",
+            DailyParams(
+                audio_out_enabled=True,
+                vad_enabled=True,
+                vad_analyzer=SileroVADAnalyzer(),
+                vad_audio_passthrough=True,
+            ),
+        )
+
+        # stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+        stt = ParakeetSTTService(api_key=os.getenv("NVIDIA_API_KEY"))
+
+        tts = FastpitchTTSService(api_key=os.getenv("NVIDIA_API_KEY"))
+
+        llm = OpenAILLMService(api_key=os.getenv("OPENAI_API_KEY"), model="gpt-4o")
+
+        messages = [
+            {
+                "role": "system",
+                "content": "You are a helpful LLM in a WebRTC call. Your goal is to demonstrate your capabilities in a succinct way. Your output will be converted to audio so don't include special characters in your answers. Respond to what the user said in a creative and helpful way.",
+            },
+        ]
+
+        context = OpenAILLMContext(messages)
+        context_aggregator = llm.create_context_aggregator(context)
+
+        pipeline = Pipeline(
+            [
+                transport.input(),  # Transport user input
+                stt,  # STT
+                context_aggregator.user(),  # User responses
+                llm,  # LLM
+                tts,  # TTS
+                transport.output(),  # Transport bot output
+                context_aggregator.assistant(),  # Assistant spoken responses
+            ]
+        )
+
+        task = PipelineTask(pipeline, PipelineParams(allow_interruptions=True))
+
+        @transport.event_handler("on_first_participant_joined")
+        async def on_first_participant_joined(transport, participant):
+            # Kick off the conversation.
+            messages.append({"role": "system", "content": "Please introduce yourself to the user."})
+            await task.queue_frames([LLMMessagesFrame(messages)])
+
+        runner = PipelineRunner()
+
+        await runner.run(task)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

src/pipecat/services/riva.py

@@ -0,0 +1,244 @@
+#
+# Copyright (c) 2024, Daily
+#
+# SPDX-License-Identifier: BSD 2-Clause License
+#
+
+import asyncio
+from typing import AsyncGenerator, List, Optional, Union, Iterator
+
+from loguru import logger
+from pydantic.main import BaseModel
+
+from pipecat.frames.frames import (
+    CancelFrame,
+    EndFrame,
+    ErrorFrame,
+    Frame,
+    InterimTranscriptionFrame,
+    StartFrame,
+    TranscriptionFrame,
+    TTSAudioRawFrame,
+    TTSStartedFrame,
+    TTSStoppedFrame,
+)
+from pipecat.services.ai_services import STTService, TTSService
+from pipecat.transcriptions.language import Language
+from pipecat.utils.time import time_now_iso8601
+
+try:
+    import riva.client
+
+except ModuleNotFoundError as e:
+    logger.error(f"Exception: {e}")
+    logger.error(
+        "In order to use nvidia riva TTS or STT, you need to `pip install pipecat-ai[riva]`. Also, set `NVIDIA_API_KEY` environment variable."
+    )
+    raise Exception(f"Missing module: {e}")
+
+
+class FastpitchTTSService(TTSService):
+    class InputParams(BaseModel):
+        language: Optional[str] = "en-US"
+
+    def __init__(
+        self,
+        *,
+        api_key: str,
+        server: str = "grpc.nvcf.nvidia.com:443",
+        voice_id: str = "English-US.Female-1",
+        sample_rate_hz: int = 24000,
+        # nvidia riva calls this 'function-id'
+        model: str = "0149dedb-2be8-4195-b9a0-e57e0e14f972",
+        params: InputParams = InputParams(),
+        **kwargs,
+    ):
+        super().__init__(sample_rate=sample_rate_hz, **kwargs)
+        self._api_key = api_key
+
+        self.set_model_name("fastpitch-hifigan-tts")
+        self.set_voice(voice_id)
+
+        self.voice_id = voice_id
+        self.sample_rate_hz = sample_rate_hz
+        self.language_code = params.language
+        self.nchannels = 1
+        self.sampwidth = 2
+        self.quality = None
+
+        metadata = [
+            ["function-id", model],
+            ["authorization", f"Bearer {api_key}"],
+        ]
+        auth = riva.client.Auth(None, True, server, metadata)
+
+        self.service = riva.client.SpeechSynthesisService(auth)
+
+    async def stop(self, frame: EndFrame):
+        await super().stop(frame)
+
+    async def cancel(self, frame: CancelFrame):
+        await super().cancel(frame)
+
+    async def run_tts(self, text: str) -> AsyncGenerator[Frame, None]:
+        logger.debug(f"Generating TTS: [{text}]")
+
+        await self.start_ttfb_metrics()
+        yield TTSStartedFrame()
+
+        try:
+            custom_dictionary_input = {}
+            responses = self.service.synthesize_online(
+                text,
+                self.voice_id,
+                self.language_code,
+                sample_rate_hz=self.sample_rate_hz,
+                audio_prompt_file=None,
+                quality=20 if self.quality is None else self.quality,
+                custom_dictionary=custom_dictionary_input,
+            )
+
+            for resp in responses:
+                await self.stop_ttfb_metrics()
+
+                frame = TTSAudioRawFrame(
+                    audio=resp.audio,
+                    sample_rate=self.sample_rate_hz,
+                    num_channels=self.nchannels,
+                )
+                yield frame
+
+        except Exception as e:
+            logger.error(f"{self} exception: {e}")
+
+        await self.start_tts_usage_metrics(text)
+        yield TTSStoppedFrame()
+
+
+class ParakeetSTTService(STTService):
+    def __init__(
+        self,
+        *,
+        api_key: str,
+        server: str = "grpc.nvcf.nvidia.com:443",
+        # nvidia calls this 'function-id'
+        model: str = "1598d209-5e27-4d3c-8079-4751568b1081",
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+        self._api_key = api_key
+
+        self.set_model_name("parakeet-ctc-1.1b-asr")
+
+        input_device = 0
+        list_devices = False
+        profanity_filter = False
+        automatic_punctuation = False
+        no_verbatim_transcripts = False
+        language_code = "en-US"
+        model_name = ""
+        boosted_lm_words = None
+        boosted_lm_score = 4.0
+        speaker_diarization = False
+        diarization_max_speakers = 3
+        start_history = -1
+        start_threshold = -1.0
+        stop_history = -1
+        stop_threshold = -1.0
+        stop_history_eou = -1
+        stop_threshold_eou = -1.0
+        custom_configuration = ""
+        ssl_cert = None
+        use_ssl = True
+        sample_rate_hz: int = 16000
+        file_streaming_chunk = 1600
+
+        metadata = [
+            ["function-id", model],
+            ["authorization", f"Bearer {api_key}"],
+        ]
+        auth = riva.client.Auth(None, True, server, metadata)
+
+        self.asr_service = riva.client.ASRService(auth)
+
+        config = riva.client.StreamingRecognitionConfig(
+            config=riva.client.RecognitionConfig(
+                encoding=riva.client.AudioEncoding.LINEAR_PCM,
+                language_code=language_code,
+                model="",
+                max_alternatives=1,
+                profanity_filter=profanity_filter,
+                enable_automatic_punctuation=automatic_punctuation,
+                verbatim_transcripts=not no_verbatim_transcripts,
+                sample_rate_hertz=sample_rate_hz,
+                audio_channel_count=1,
+            ),
+            interim_results=True,
+        )
+        self.config = config
+        riva.client.add_word_boosting_to_config(config, boosted_lm_words, boosted_lm_score)
+        riva.client.add_endpoint_parameters_to_config(
+            config,
+            start_history,
+            start_threshold,
+            stop_history,
+            stop_history_eou,
+            stop_threshold,
+            stop_threshold_eou,
+        )
+        riva.client.add_custom_configuration_to_config(config, custom_configuration)
+
+        # this doesn't work, but something like this perhaps? part 1
+        self.audio = []
+        self.responses = self.asr_service.streaming_response_generator(
+            audio_chunks=[self.audio],
+            streaming_config=self.config,
+        )
+
+    def can_generate_metrics(self) -> bool:
+        return False
+
+    async def start(self, frame: StartFrame):
+        await super().start(frame)
+
+    async def stop(self, frame: EndFrame):
+        await super().stop(frame)
+
+    async def cancel(self, frame: CancelFrame):
+        await super().cancel(frame)
+
+    async def run_stt(self, audio: bytes) -> AsyncGenerator[Frame, None]:
+        # this doesn't work, but something like this perhaps? part 2
+        self.audio.append(audio)
+
+        # need to start to run this generator only once somewhere...
+        # 'start' function doesn't work...
+        # something about the event loop...
+        # maybe an audio buffer... though my attempt at that didn't work either
+        for response in self.responses:
+            if not response.results:
+                continue
+            partial_transcript = ""
+            for result in response.results:
+                if result:
+                    if not result.alternatives:
+                        continue
+                    transcript = result.alternatives[0].transcript
+                    if transcript:
+                        language = None
+                        if len(transcript) > 0:
+                            await self.stop_ttfb_metrics()
+                            if result.is_final:
+                                await self.stop_processing_metrics()
+                                yield TranscriptionFrame(
+                                    transcript, "", time_now_iso8601(), language
+                                )
+                            else:
+                                yield InterimTranscriptionFrame(
+                                    transcript, "", time_now_iso8601(), language
+                                )
+        yield None
+
+    async def _on_speech_started(self, *args, **kwargs):
+        await self.start_ttfb_metrics()
+        await self.start_processing_metrics()