feat: add clone-voice-10s demo - #27
Open
luke-speechify wants to merge 3 commits into
Open
Conversation
Zero-shot voice cloning from a ~10 second sample with an explicit consent gate, immediate synthesis with the clone, and auto-delete afterwards. Three Node-runtime routes (clone, speak, delete) hold the Speechify key server-side and are gated with Turnstile. Registered as a Vercel Service under /clone-voice-10s.
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
|
Deployment failed for project speechify-api-demos with the following error: Learn More: https://vercel.com/docs/concepts/projects/project-configuration |
|
Deployment failed for project speechify-api-demos with the following error: Learn More: https://vercel.com/docs/concepts/projects/project-configuration |
Vercel service names must match ^[a-z]([a-z_-]*[a-z])?$ (no digits), so the service key clone-voice-10s failed vercel.json schema validation and the preview never built. Rename the service to clone-voice-ten-s; the slug, URL path (/clone-voice-10s), and folder root are unchanged.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Pairs with the upcoming speechify.ai post "Clone a voice from 10 seconds and ship it today".
What it does
A one-page Next.js demo for zero-shot voice cloning from a short (~10 second) sample, with consent as a first-class step:
fixtures/spacewalk.wav(public-domain NASA audio) so you can run the whole flow without recording anything.consent: JSON.stringify({ fullName, email }).How it works
nodejs-runtime routes holdSPEECHIFY_API_KEYserver-side, never exposing it to the browser:POST /api/clone(client.voices.create, catchesSpeechifyError402 with a friendly "not on your plan" message),POST /api/speak(client.audio.speech,simba-english), andDELETE /api/voice?id=…(client.voices.delete).verifyTurnstile), and the client attaches anx-turnstile-tokenheader to every gated fetch.@speechify/apiis markedserverExternalPackagesso the SDK never enters the client bundle.Hosting registration
vercel.json— added aclone-voice-10sservice and a/clone-voice-10s/:path*rewrite (before the/and catch-all rewrites).pnpm-workspace.yaml— addeddemos/clone-voice-10s.pnpm generateto refresh the README table and site manifest. Dev port 8772.Verification
pnpm --filter clone-voice-10s build✅simba-english) → delete flow against the real API with the shippedspacewalk.wav; all three steps returned successfully. The 402 path returns a clean plan message when cloning is not on the key's plan.