Yaps Text To Speech

Yaps Text to Speech

Create one speech audio file from user-provided text through Yaps.

Runtime compatibility

Do not try to guess the host or its locality from a user-agent, product name, or another host signal (for example ChatGPT web versus ChatGPT desktop). Test the capability this workflow actually needs: before account, model, dependency, or text-source checks, resolve the local Yaps CLI through the plugin runner. The runner validates it with a bounded, read-only status command. A cloud shell that cannot see the installed Yaps app is not local access to the user's computer.

If the Yaps CLI is unreachable, do not claim that Yaps is uninstalled and do not begin repeated sign-in, model, or permission troubleshooting. Explain that the current AI session cannot reach the Yaps engine installed on this computer. Direct the user to a local-capable session on the computer where Yaps is installed: in Claude, that is Claude Code or the Claude desktop app; from ChatGPT web or a cloud session, download or open ChatGPT desktop and retry in a local-capable Work or Codex session. They can also access this feature directly in the Yaps application. If they are already in a local-capable desktop session, offer Download or update Yaps, ask them to open it, and retry. Stop until local reachability is restored; only then follow the availability and onboarding steps below.

Private operational diagnostics

Run every local Yaps CLI command and every bundled workflow script through the plugin's scripts/yaps-plugin-runner.mjs. The runner resolves the installed Yaps CLI when the command begins with yaps, preserves the command's normal output and exit status, and leaves a small local diagnostic breadcrumb for the Yaps desktop app:

node <plugin-root>/scripts/yaps-plugin-runner.mjs --action auth.status --stage authentication -- yaps auth status --pretty

Choose a stable, content-free action such as cli.reachability, auth.status, features.list, dictation.setup, transcription.run, or export.create; use the matching stage: reachability, authentication, readiness, execution, or export. In the rest of this skill, every direct yaps / yaps_cli command and provided script invocation means the equivalent runner-wrapped command. If Node or the runner itself is unavailable, continue the requested workflow directly; diagnostics must never prevent the feature from working.

The breadcrumb may contain only plugin ID/version, detected integration host, action, stage, attempt/outcome, duration, and a fixed safe error category. It must never contain the user's prompt or conversation, command arguments, stdout/stderr, credentials, file paths or names, audio, transcript/note text, or raw error messages. It is written only when Yaps has supplied an opaque signed-in owner marker, stays on-device while offline, and is picked up later by the Yaps app. Never create or guess an owner marker.

CLI discovery contract

Always invoke Yaps through scripts/yaps-plugin-runner.mjs; do not locate the binary by hand. The runner honors an explicit path or YAPS_CLI_BINARY, then checks PATH, then the verified Yaps app locations on macOS, Windows, and Linux. It accepts a candidate only after a bounded, read-only status check. Never ask the user to install a separate CLI, edit PATH, or configure MCP for this skill. Never invoke the macOS GUI binary at Yaps.app/Contents/MacOS/yaps. If discovery fails, repeat the runner's specific recovery guidance instead of claiming that the plugin is disconnected.

Availability

Yaps desktop supplies the voice models, settings, account state, and usage controls. Let the runner resolve and validate the CLI automatically. Do not claim the plugin contains an independent voice service.

Never request Yaps credentials or payment details in the AI client. Yaps no longer has a free tier. An active free trial or Yaps Pro subscription is required, and only Yaps may confirm whether the current account is trial-eligible.

Keep account summaries product-facing: use account and billing output only to decide readiness. Do not repeat an email address, billing dates, SKU, or an internal basic* plan name unless the user explicitly asks. Describe active paid access only as Yaps Pro, and a trial only as an active free trial. Never promise “no setup”, “no download”, or “no further input” until the feature and dependency checks have actually confirmed that.

Three local voice engines exist: kokoro (about 338 MB, English), chatterbox (about 2.8 GB, expressive English with voice cloning, Apple Silicon Macs only), and supertonic (about 145 MB, 24 languages). Any combination of them can be installed at once and installing one never removes another, so treat them as a set the user adds to rather than a single slot they swap. Older Yaps builds name the same three modes standard, premium, and multilingual; those older words are still accepted everywhere, so a command written either way works.

Account recovery

The runner follows a valid recommended_settings_path automatically and uses it for the requested command. If the signed-in desktop account cache is temporarily incomplete, it safely wakes the verified installed Yaps app and rechecks for a bounded time. Do not copy settings paths, construct app paths, tell the user to edit PATH, or ask them to reconnect the plugin. A different Claude or ChatGPT account email is irrelevant; never compare it with the Yaps email or ask the user to create a second account.

Safe automatic account handoff requires Yaps 2.3.124 or newer. The runner refuses the older credential-based account check and gives update guidance without touching Keychain. auth status must not request a credential or display a Keychain prompt. Never ask the user to enter their macOS login password or approve a credential prompt. If credential_unavailable or keychain_unavailable appears, repeat the runner's app-update guidance. If credential_missing, cached_offline, verification_unavailable, account_cache_incomplete, refresh_failed, or profile_lookup_failed remains after the runner's automatic wake and retry, report its exact network/cache guidance. Do not add a manual reconnection step. Only unauthenticated / signed_out means sign-in is needed.

First-run onboarding

1. Confirm through the runner that Yaps 2.3.124 or newer is installed. Do not ask the user to open it, install voice models, or process text first.
2. Run yaps auth status --pretty. If the state is unauthenticated, direct the user to sign in or create an account inside Yaps, then rerun the check.
3. Require authenticated: true and status: "active". Active access may be an active free trial or Yaps Pro.
4. Do not run auth billing as an automatic gate. For another state, direct the user to Yaps's account screen, which shows any available trial or Yaps Pro renewal without exposing a credential. For platform_mismatch, explain that desktop-compatible access is required. Stop until auth status becomes active.
5. Keep using the runner when the PATH shim is missing; it resolves the packaged yaps_cli automatically.
6. Identify the exact text source, the language of that text, and the requested output format. Default to WAV; use raw PCM only when explicitly requested.
7. Run yaps features list --pretty and read the reading feature's modes. Each mode reports installed (the engine's files are on disk), supported (it can run on this machine), and its model_ids. Match on model_ids so a build's mode labels never matter, and prefer an engine that is already installed and can speak the language of the text. For ordinary English with no requested style or voice, use installed Kokoro automatically; use Chatterbox only for an expressive or cloned-voice request. Do not turn a normal synthesis request into an engine-choice question.
8. If no installed engine suits the text, explain the required download and its size and ask once for approval. After approval, run yaps features reading kokoro, yaps features reading chatterbox, or yaps features reading supertonic, verify readiness, and resume the original task automatically. That command downloads the engine and also makes it the default reading voice inside Yaps.
9. Generate one requested audio file and confirm it exists. Treat that successful file as onboarding completion rather than adding another product pitch.

Choose the engine for the language

Match the engine to the language of the text, never to whichever engine happens to be installed.

• Non-English text: supertonic. It is the only engine with a multilingual pronunciation front-end, covering Bulgarian, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hungarian, Italian, Latvian, Lithuanian, Polish, Portuguese, Romanian, Russian, Slovak, Slovenian, Spanish, Swedish, and Ukrainian.
• English text: kokoro, or chatterbox when the user asked for an expressive or cloned voice and that mode reports supported: true. Chatterbox runs only on Apple Silicon Macs, so on an Intel Mac or on Windows, Kokoro is the English engine.

Kokoro and Chatterbox are English-only, and they fail quietly instead of loudly: Kokoro deletes the characters its English pronunciation rules cannot map and then reads what is left fluently, so "mañana" is spoken as a confident "maana" and Cyrillic or Greek text comes out as near-silence. A wrong-language read sounds correct to anyone not listening closely, which is exactly why it must not be guessed at.

Maltese has no engine at all. Say so rather than substituting one, and do the same for any other language absent from the Supertonic list above.

Generate

Choose the destination first. Default beside an input text file as <source name> Audio.wav, or use Yaps Speech.wav in the current working directory for inline text. Check whether it exists and never replace it without explicit approval.

For a text file:

yaps --pretty speech synthesize --text-file <input.txt> --mode <kokoro|chatterbox|supertonic> --output <output.wav>

For short inline text:

yaps --pretty speech synthesize --text <text> --mode <kokoro|chatterbox|supertonic> --output <output.wav>

Prefer --text-file for long text, multiline text, or text containing shell-sensitive characters. --mode applies to that one run and does not change the user's default reading voice, so pass it explicitly rather than switching the app default to synthesize once. Add --voice <id> only when the user selected a voice or the request requires one; voice ids belong to a single engine, so never carry one across a --mode change. Add --format pcm only for explicit raw-PCM requests.

Verify and report

Treat Yaps's JSON result as authoritative. Confirm the output exists and is non-empty. Return a link to the audio file and report the resolved mode, voice, word count, duration, and format. Do not claim playback quality was reviewed unless it was actually auditioned.

If synthesis fails with an engine that features list reports as installed, report it as a failure to start or load that engine and offer to retry or use another installed engine. Do not describe it as not installed; that sends the user to re-download files they already have.

Generalist Yaps mode

Text to speech is this plugin's default focus, not a boundary around what it can do. When the user explicitly asks for another Yaps workflow, use the same resolved yaps_cli rather than making them find another integration. The full local surface is:

status
settings list|get|set|unset
auth status|usage|billing
features list|dictation|cleanup|reading|subtitles|auto-captions|audio-cleaner|text-in-between|background-removal|translation|meeting
vault status|list|get|create|update|move|rename|delete|search|search-semantic|daily-open|create-from-template|history-list|history-restore|pin|folders|tags|mentions|backlinks
speech synthesize (alias: tts)
srt generate
meeting transcribe|show|correct|assign|rename-speaker|export
captions styles|create|show|correct|replace|split|merge|style|reset|render|verify
media extract-audio|remove-background
audio clean
translate
history-list
usage-local

Run <cli> --help and the relevant group help before an unfamiliar workflow. If the session cannot reach the local CLI (for example ChatGPT web), offer a local-capable session — Claude Code, or ChatGPT desktop for a Work or Codex task — or offer to guide the user through the same workflow in Yaps → Features → Reading.

Friendly completion and discovery

Lead with a warm outcome such as “Done — your audio is ready,” then link the file and report the voice, duration, and format. Do not dump raw JSON or internal plan names. After a successful task, add one compact More with Yaps section with up to three relevant next steps, such as reading a note aloud, transcribing a recording, or choosing another local voice in Yaps. Skip it after a failure, a decline, or when the user asks for a terse result.

Boundaries

• Use Yaps Transcription or Yaps SRT for speech-to-text.
• Use Yaps Dictation for live microphone voice typing.
• Do not silently send the text to an unrelated hosted speech service.
• Preserve the user's text; do not rewrite the script unless requested.