alicewuv/kitten-tts 🔢📝❓ → 🖼️

▶️ 24 runs 📅 Aug 2025 ⚙️ Cog 0.16.2
audio cpu-friendly real-time text-to-speech tts voice-synthesis

About

Lightweight text to speech models

Example Output

Output

Example output

Performance Metrics

12.36s Prediction Time
20.96s Total Time
All Input Parameters 
{
  "text": "Hello from Kitten TTS!",
  "speed": 1,
  "format": "wav",
  "voice_type": "auto",
  "sample_rate": 24000
}
Input Parameters
seed Type: integer
Random seed (best effort only)
text Type: stringDefault: Hello from Kitten TTS!
Text to synthesize
speed Type: numberDefault: 1Range: 0.5 - 2
Speech speed (0.5–2.0)
voice Type: string
Exact voice name (optional). Overrides voice_type.
format Default: wav
Output format
voice_type Default: auto
Voice type selector
sample_rate Type: integerDefault: 24000
Force output sample rate (Hz)
Output Schema

Output

Type: string • Format: uri

Example Execution Logs 
Audio saved to /tmp/tmpyp884x63/audio.wav
Version Details
Version ID
860a2de5199547a7b3616abc34bee2239bf5c620c30d503e5e66edaba264b6b8
Version Created
August 12, 2025
Run on Replicate →