elevenlabs-tts_skill

This skill helps you generate high-quality audio from text using Eleven Labs, enabling podcasts, narration, and voice-overs with customizable voices.

Python

3

GitHub Stars

1

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 krishagel/geoffrey --skill elevenlabs-tts

SKILL.md3.8 KB

Overview

This skill generates high-quality audio from text using the Eleven Labs API for podcasts, narration, voice-overs, and audio summaries. It provides ready-to-use scripts to produce MP3 files, auto-chunk long texts, and return metadata about each generation. The default voice and model settings are tuned for long-form narration while allowing quick overrides per-run.

How this skill works

Run the provided CLI scripts to convert either inline text or text files into MP3 audio. The generator selects a voice and model, auto-chunks input exceeding model character limits at sentence boundaries, synthesizes each chunk, and concatenates chunks into a single MP3. The script returns JSON metadata including file path, voice, model, character count, and chunk count.

When to use it

Create podcast segments or daily audio newsletters from written summaries
Produce narrated reports, training modules, or explainer videos
Generate voice-overs for marketing videos and social media content
Quickly test voice options and output using the CLI for small scripts or demos
Convert long documents into a single, continuous MP3 via automatic chunking

Best practices

Choose a model that matches length and expression needs (multilingual v2 for long-form, flash/turbo for larger limits, v3 for expressive short clips)
Respect per-model character limits; rely on auto-chunking for long inputs but split semantically for best pacing
Pick curated voices by use case (Rachel for narration, Josh for news, Bella for calm storytelling)
Provide clean, edited text to avoid awkward pauses or mispronunciations
Run list_voices.py to discover available and premade voices before bulk generation

Example use cases

Daily summary podcast: generate separate segments (weather, news, calendar) and concatenate into an episode
Long report narration: auto-chunk a quarterly report and output a single MP3 for stakeholders
Tutorial voiceover: create step-by-step audio for instructional videos using an expressive voice
Quick QA: run a short text snippet locally to verify voice, pacing, and model selection

FAQ

Output is MP3 only; use external tools to convert formats if needed.

Can I use multiple voices in a single output?

Not in one run — the tool supports one voice per generated file. Generate separate files per voice and concatenate externally if required.