Downloads · 30 days
0
PersonaFlow
Kailing-Leifang/PersonaFlow
PersonaFlow is a machine learning model from Kailing-Leifang. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Interactive Audio Character Demo - Speak with AI characters that have distinct personalities, voices, and animated portraits.
Access
Public
Updated Jan 23, 2026
Repo size
—
Likes
0
Public
Repo makeup
Click a slice to open those files.
.py8.9 KB · 73%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Jan 23, 2026
- Updated
- Jan 23, 2026
- SHA
- d35a6a0e
Model card
From the Hugging Face model README
🎭 PersonaFlow
Interactive Audio Character Demo - Speak with AI characters that have distinct personalities, voices, and animated portraits.
Features
- Voice Input: Speak into your microphone (up to 10 seconds)
- Multiple Characters: Choose from 3 distinct personalities
- 🚀 The Visionary - Bold, dramatic, futuristic
- 🤔 The Skeptic - Dry, questioning, sardonic
- 🌟 The Guide - Warm, helpful, encouraging
- Unique Voices: Each character has a distinct voice
- Animated Portraits: Visual feedback with lip-sync animation
- Conversation History: Track your dialogue with each character
How It Works
- Select a character from the dropdown
- Click the microphone and speak your message
- Listen to the character's response with their unique voice
- Continue the conversation or switch characters
Technology
- STT: distil-whisper/distil-large-v3 (faster-whisper backend)
- LLM: Qwen/Qwen2.5-3B-Instruct
- TTS: Kokoro-82M with multiple voice options
- Frontend: Gradio with custom CSS animations
Latency Target
| Component | Target |
|---|---|
| STT | <500ms |
| LLM | <400ms |
| TTS | <300ms |
| Total | <1.5s |
Built for engaging 1-2 minute voice interactions with AI personalities.
- License
- apache-2.0
- Created
- Jan 23, 2026
- Updated
- Jan 23, 2026