agentleFS
MDizzleZAMDizzleZA/claude-skills/fish-audio-tts/SKILL.md

fish-audio-tts

Skill0 starsChanged 3 months agoMITView on GitHub

Text-to-speech via the Fish Audio API — free tier (s2.1-pro-free model) with voice cloning support. Use whenever speech audio, a voiceover, narration, or an audio version of text is needed — video narration, explainer voiceovers, audio ads, IVR/phone prompts, podcast snippets, or converting a script/blog post to MP3/WAV. API key already stored in Credential Manager; generation is free on the default model.

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 3

    1. calesthio/OpenMontage★ 81ksame name
    2. OpenMontageInc/OpenMontage★ 81ksame name
    3. Gopyr/gopyr-skills-hub★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Default (no reference_id): a generic voice — fine for drafts and internal use.

    line 43

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Fish Audio TTS

Generate speech from text via https://api.fish.audio/v1/tts. The free s2.1-pro-free model is the default — no cost per call. API key is in Windows Credential Manager (fish-audio / apikey); the script reads it automatically.

Quick start

python ~/.claude/skills/fish-audio-tts/scripts/tts.py "Hello, world!" -o output.mp3
python ~/.claude/skills/fish-audio-tts/scripts/tts.py --file script.txt -o narration.mp3

Options: --voice <reference_id> (voice from fish.audio library or a cloned voice), --format mp3|wav|pcm|opus, --speed 0.5-2.0, --bitrate 64|128|192, --model s2.1-pro (paid, higher quality), --no-normalize (skip number/date expansion).

For long scripts prefer --file; the API chunks text automatically (chunk_length default 200 chars).

Direct API (when the script doesn't fit)

import httpx, keyring

res = httpx.post(
    "https://api.fish.audio/v1/tts",
    headers={
        "Authorization": f"Bearer {keyring.get_password('fish-audio', 'apikey')}",
        "Content-Type": "application/json",
        "model": "s2.1-pro-free",          # the free tier lives in this HEADER, not the body
    },
    json={"text": "Hello, world!", "format": "mp3"},
    timeout=300,
)
open("output.mp3", "wb").write(res.content)

Key body fields: text, reference_id (voice), format, mp3_bitrate, sample_rate, prosody: {speed, volume}, normalize, latency: "balanced"|"normal", chunk_length (100–300). Models (header): s2.1-pro-free (free), s2.1-pro (production), s2-pro, s1 (supports emotion tags). Full docs: https://docs.fish.audio/features/text-to-speech

Voices

  • Default (no reference_id): a generic voice — fine for drafts and internal use.
  • Browse voices at https://fish.audio (Voice Library) and pass the model ID as --voice.
  • Instant cloning: pass references (audio sample + transcript) in the body instead of reference_id — use msgpack (Content-Type: application/msgpack) to send raw audio bytes. Only clone voices with the owner's permission.

SA / client-work notes

  • Listen before delivering: check pronunciation of SA names and terms (e.g. "Gauteng", client names); tweak spelling phonetically in the input text if mispronounced.
  • Financial-services (FSP) audio content is advertising under FAIS — run scripts through compliance review before generating final audio.
  • For video narration workflows (hyperframes/faceless-explainer), the media-use HeyGen path is the default; use this skill when a free/simple standalone voiceover file is all that's needed, then adopt the file as an asset.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.