[Infra] Piper TTS Voice Model — Train Timmy's Custom Voice #853

Closed
opened 2026-03-21 22:21:07 +00:00 by perplexity · 0 comments
Collaborator

Scope

Train a custom Piper TTS voice model for Timmy so he has a consistent, recognizable voice across all content (journal narration, stream audio, daily episodes).

Requirements

  • Select or record voice training data that fits Timmy's character
  • Train Piper TTS model (VITS architecture)
  • Integrate with FastAPI harness for on-demand TTS generation
  • Low latency: suitable for real-time stream narration
  • Voice should feel natural, not robotic — Timmy is a character, not a chatbot

Used By

  • Phase 5 stream audio
  • Phase 7 daily episode narration
  • AlexanderWhitestone.com /tower interactions

Assignee: Kimi

## Scope Train a custom Piper TTS voice model for Timmy so he has a consistent, recognizable voice across all content (journal narration, stream audio, daily episodes). ## Requirements - Select or record voice training data that fits Timmy's character - Train Piper TTS model (VITS architecture) - Integrate with FastAPI harness for on-demand TTS generation - Low latency: suitable for real-time stream narration - Voice should feel natural, not robotic — Timmy is a character, not a chatbot ## Used By - Phase 5 stream audio - Phase 7 daily episode narration - AlexanderWhitestone.com /tower interactions ## Assignee: Kimi
gemini was assigned by Rockachopa 2026-03-22 23:34:06 +00:00
claude added the harnessp2-backlog labels 2026-03-23 13:56:11 +00:00
gemini was unassigned by Timmy 2026-03-24 19:33:57 +00:00
Timmy closed this issue 2026-03-24 21:54:51 +00:00
Sign in to join this conversation.
No Label harness p2-backlog
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Rockachopa/Timmy-time-dashboard#853