sapi-ttsr_skill

This skill enables lightweight Windows SAPI5 text-to-speech in Python, auto-selecting voices and producing instantly playable WAV files.

Python

1.1k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill sapi-ttsr

_meta.json282 B
SKILL.md6.5 KB

Overview

This skill provides lightweight Windows SAPI5 text-to-speech using system voices, including Windows 11 Neural voices. It produces instant WAV output with zero GPU usage and auto-selects the best voice for the requested language. It works on Windows 10 and 11 and is designed for fast, low-resource TTS needs.

How this skill works

The skill uses the System.Speech.Synthesis API to enumerate installed SAPI5 voices, prefer Neural voices when available, and synthesize text to a WAV file. It auto-selects a voice based on language or a provided partial voice name, allows rate adjustments, and can play the generated audio immediately. If no output path is supplied, it writes files into a user workspace with timestamped names.

When to use it

Generate spoken audio quickly without GPU or heavy dependencies.
Add TTS to Windows-based scripts, chatbots, or automation workflows.
Produce multilingual audio by auto-selecting the best voice for a language.
Create temporary WAV files for previews, demos, or voice notifications.
Run on machines where installing large TTS models or GPU access is impractical.

Best practices

Use -ListVoices first to inspect available system voices and confirm language support.
Prefer Neural voices on Windows 11 for higher quality; install extra voices via Windows Settings if needed.
Provide a Language code (en, fr, de, es, it) to improve voice selection when no specific name is given.
Adjust Rate between -10 and +10 for natural pacing; test with short phrases before batch processing.
Store outputs in a dedicated workspace to avoid clutter and enable easy cleanup.

Example use cases

Create quick spoken prompts for a desktop chatbot using system voices.
Batch-generate notification sounds for automation scripts without external dependencies.
Produce localized audio snippets for demos by specifying -Lang en|fr|de|etc.
Preview how a UI will sound by calling the script with -Play to auto-play results.
Provide fallback TTS on devices that cannot run neural models locally (no GPU).

FAQ

No. It uses Windows SAPI5 and requires no GPU; generation is effectively instant.

How do I get better voices?

Install Neural voices via Settings → Time & Language → Speech on Windows 11, or use adapter tools to expose additional voices on Windows 10/11.