Your First Yaazh Generate: Start With a Built-In Voice
You do not need a clone, an account, or a credit card to hear Yaazh speak. The fastest honest path is Generate with a built-in voice: Adam, Bella, or Heart in English (Kokoro), or Rasa in Tamil (Piper).
This walkthrough is educate-first. It stays inside published product language, softens any purchase talk, and reminds you that the free hosted studio is a shared preview—great for learning, not for precious keepsakes.
Before you start: two ground rules
- Permission culture — If you later clone, only use voices you have permission to keep. A clone is not a replacement for a person. For this first run, skip cloning entirely.
- Shared preview honesty — At app.yaazh.nexageapps.com, many people share a server. Demo storage may reset. Fine for practice lines; not for the only copy of a family sample.
Mac permanence is Coming soon (beta). Today’s goal is simply: type a line, hear speech, understand the studio shape.
Step 1 — Open the studio (no account)
Go to the free hosted studio: https://app.yaazh.nexageapps.com/.
Yaazh’s motto is No account · No ads · No server · Privacy first. You should not be gated behind a Yaazh login to try Generate with built-ins. If something asks you to invent an account that the brand does not describe, stop and re-check the official site.
Step 2 — Find Generate
Look for Generate among the local-first surfaces: Generate / Voices / Studio / Effects / Live / Agents / Share.
Generate is the “type → pick a voice → hear them” lane. You can explore Studio and Effects later; first success beats feature tourism.
Step 3 — Pick a built-in voice
Start with one of these published built-ins:
- Adam — default English Generate path via Kokoro
- Bella — warm English Kokoro speaker
- Heart — soft English Kokoro speaker
- Rasa — Tamil Piper voice (a real female speaker per product copy)
New sessions are designed to default toward Adam · English · Kokoro so you are less likely to hit a “voice missing” empty state. If you want Tamil, switch to Rasa deliberately.
Avoid empty custom clones for this first run. Empty clones cannot generate—another reason built-ins are the kinder onboarding.
Step 4 — Write one short line
Keep the first line boring on purpose:
- English: “Drive safe. Call when you reach.”
- Tamil (if using Rasa): a short phrase you are comfortable testing in a shared preview
Why short? You learn latency, loudness, and whether Effects interest you—without dumping a private journal into a demo server.
Step 5 — Generate and listen
Run Generate. Listen once without tweaking. Then optionally:
- Try the same line on Bella or Heart to hear Kokoro personality differences
- Switch to Rasa for a Tamil take
- Preview an Effect pack (Phone call, Warm hug, Radio host, Storytime, Temple hall, Night whisper, plus classics) dry/wet before you save anything you care about
You are still in learning mode. Saving precious audio in the shared preview is optional and risky if you need it later.
Step 6 — Share the right way (optional)
If you export a Story or Reel vibe later, remember: Yaazh never auto-posts Instagram. Export locally, copy captions if offered (Tamil or English), then open Instagram yourself.
That constraint is a feature for consent, not a missing integration.
What to try after your first line
Once one Generate works:
- Studio — arrange multi-voice narrative lines on a timeline (still prefer built-ins at first)
- Live — speak in the studio; on-device transcription is part of the Mac-oriented story
-
Agents — assign a voice and Preview (advanced users may later touch optional MCP such as
voicebox.speakatapp.yaazh.nexageapps.com/mcp—skip until you need it) - Voices — only then consider creating a voice from a clip you have permission to keep—and prefer Mac when permanence matters
Common first-run mistakes
- Cloning Mum/Friend/You samples into the free preview as the only copy
- Expecting private cloud retention without an account
- Hunting a “live” Mac DMG instead of joining the beta waitlist
- Assuming guaranteed clone quality (Yaazh does not claim that)
- Looking for auto Instagram posting
Soft CTA
When you are ready, open Generate in the hosted studio, start with Adam or Rasa, and keep the first line short. Read the fuller story on yaazh.nexageapps.com, skim Pricing if Mac permanence interests you, and join the Mac beta for an email when Download is truly ready. Support: tally.so/r/68qaRJ.
No pressure to unlock anything for a first listen. Hear one line. Keep permission sacred for anything that follows.
FAQ
Do I need to clone a voice to use Generate?
No. Adam, Bella, Heart, and Rasa are enough for a first success.
Why does my custom voice fail?
Empty clones cannot generate. Use a real permitted sample—or stick to built-ins.
Will my practice audio stay forever in the browser studio?
Not promised. Shared preview demo storage may reset anytime.
Can I generate in Tamil?
Yes—start with built-in Rasa (Piper).
Does Generate require MCP?
No. MCP (voicebox.speak) is an advanced optional path for agents.
