Shorts are under 60 seconds. You need fast, attention-grabbing narration. Kitten TTS with speed=1.3-1.5 is perfect.

Why Kitten TTS for Shorts?

  • Quick generation (under 1 second per clip)
  • 8 voices for variety across different Shorts
  • Speed control for fast-paced delivery
  • Batch processing for daily content
  • Free -- no per-video costs

Example: Fast Shorts Script

from kittentts import KittenTTS
model = KittenTTS("KittenML/kitten-tts-micro-0.8")

# Shorts need faster pace
model.generate_to_file(
    "This one trick will change how you edit videos forever!",
    "hook.wav", voice="Kiki", speed=1.3
)

model.generate_to_file(
    "Most editors waste hours on this. Here is the fix in 30 seconds.",
    "body.wav", voice="Jasper", speed=1.4
)

model.generate_to_file(
    "Follow for more editing tips! Drop a comment if this helped.",
    "cta.wav", voice="Rosie", speed=1.3
)

Best Practices for Shorts

  • Hook first: Start with the most engaging line in the first 3 seconds
  • Fast delivery: Use speed 1.3-1.5 for Shorts pacing
  • Kiki or Rosie: Best voices for high-energy short-form content
  • Caption sync: Generate audio first, then match captions to timing

Batch Shorts Factory

# Generate 10 Shorts voiceovers at once
shorts_scripts = [
    ("hook", "Did you know this about space?", "Kiki", 1.4),
    ("body", "Galaxies are moving away from us faster than light.", "Jasper", 1.3),
    ("cta", "Follow for daily space facts!", "Rosie", 1.3),
]

for part, text, voice, spd in shorts_scripts:
    model.generate_to_file(text, part + ".wav", voice=voice, speed=spd)