prosody-say 0.1.4

A prosody processing library for TTS synthesis.

prosody_synthesizer

A prosody processing library for TTS synthesis on macOS that computes dynamic pitch, rate, and volume adjustments based on sentiment and energy analysis. This library relies on the system say command, which is available only on macOS.

---

Features

• Sentiment and Energy Analysis: Uses NLTK’s VADER and basic lexical heuristics to derive a sentence‐level “energy” and sentiment (compound) score.
• Dynamic Shifts: Adjusts speech rate and pitch dynamically within each phrase.
• macOS say Integration: Outputs text with inline TTS commands (e.g. [[rate]], [[pbas]]), then invokes say.
• Object-Oriented Design: Encapsulated in a ProsodySynthesizer class for easy usage.

---

Requirements

1. macOS
   This library uses the built-in say command, so it won't work on Windows or Linux unless you adapt those TTS calls.

2. Python <3.12 (recommended)

   • Although no strict upper-bound is enforced, note that spaCy may not yet support Python 3.13.
   • If you encounter issues on future Python versions, consider using 3.6–3.12.

3. Dependencies

   • nltk
   • numpy
   • spacy

Required Data Downloads

Download Required Resources

Before using prosody_say, ensure that the necessary NLTK and spaCy resources are installed. You can do this by running:

prosody-download-resources

Installation

1. Install from PyPI :

   pip install prosody_synthesizer
   

2. **Install from source**:
   ```git clone https://github.com/yourusername/prosody_synthesizer.git
   cd prosody_synthesizer
   pip install .
   ```