NVIDIA FastPitch

​

Overview

FastPitchTTSService converts text to speech using NVIDIA’s Riva FastPitch TTS model. It provides high-quality text-to-speech synthesis with configurable voice options.

​

Installation

To use FastPitchTTSService, install the required dependencies:

pip install pipecat-ai[riva]

You’ll also need to set up your NVIDIA API key as an environment variable: NVIDIA_API_KEY

​

Configuration

​

Constructor Parameters

​

InputParams

​

Input

The service accepts text input through its TTS pipeline.

​

Output Frames

​

TTSStartedFrame

Signals the start of audio generation.

​

TTSAudioRawFrame

Contains generated audio data:

​

TTSStoppedFrame

Signals the completion of audio generation.

​

Methods

See the TTS base class methods for additional functionality.

​

Language Support

FastPitch TTS primarily supports English with various regional accents:

​

Usage Example

from pipecat.services.riva import FastPitchTTSService
from pipecat.transcriptions.language import Language

# Configure service
tts = FastPitchTTSService(
    api_key="your-nvidia-api-key",
    voice_id="English-US.Female-1",
    params=FastPitchTTSService.InputParams(
        language=Language.EN_US,
        quality=20
    )
)

# Use in pipeline
pipeline = Pipeline([
    ...,
    llm,
    tts,
    transport.output(),
])

​

Frame Flow

​

Metrics Support

The service supports metrics collection:

• Time to First Byte (TTFB)
• TTS usage metrics
• Processing duration

​

Audio Processing

• Processes audio through the Riva API
• Generates mono audio output
• Handles asynchronous audio streaming
• Configurable sampling rate

​

Notes

• Uses NVIDIA’s Riva AI Services platform
• Streams audio in chunks
• Requires valid NVIDIA API key
• Thread-safe processing with asyncio