[Showoff Saturday] I built a lightweight, open-source web UI and REST API for Kokoro TTS with auto-synced SRT export

Hey devs,

For the past couple of weeks, I’ve been working on a lightweight web interface and REST API wrapper around the open-source Kokoro TTS model called Kokoro Voice Studio (SayTTS).

Most hosted TTS options come with heavy subscription paywalls or complex Docker/Python CLI setups. I wanted something fast, browser-ready, and developer-friendly that generates natural audio with zero friction.

Tech highlights & features:

• Zero-auth REST API: Simple endpoints to generate audio buffers or download links for scripts and automation workflows.
• Client-side Subtitle Alignment: Generates and syncs timestamped .srt subtitle files alongside the generated .wav/.mp3 streams.
• Audio Pipeline: Built-in speed modulators and basic web-audio equalization mastering before export.
• Multi-language support: 60+ neural voice variants across English, Japanese, French, Spanish, Hindi, etc.
• No login / Free to use: Open access without auth barriers.
• Live Web App:https://saytts.site
• GitHub Repo:https://github.com/dilshan916/kokoro-voice-studio

I’d love feedback on UI responsiveness, audio pipeline stability, or suggestions for the API architecture. Let me know what you think!