Kitten TTS by KittenML is the voice engine powering this workflow. Free, open-source, offline capable.
Pipeline Overview
Step 1: Topic Research
Select episode topic. Research key points and talking points. Outline episode structure.
Step 2: Script Writing
Write host monologue, interview questions, segment transitions. Multi-speaker script if needed.
Step 3: Voice Generation
Kitten TTS: Jasper for host, Bruno for guest, Bella for co-host. Generate all audio segments.
Step 4: Audio Production
Import into Audacity/Reaper. Add intro/outro music. Mix levels. Add sound effects.
Step 5: Mastering
Normalize audio to -16 LUFS. Apply compression. Export as MP3 192kbps.
Step 6: Distribution
Upload to Spotify, Apple Podcasts, YouTube. Publish show notes and transcript.
Key Tools in This Pipeline
- Kitten TTS: Voice generation (free, offline, 8 voices)
- Python: Automation and batch processing
- Video Editor: Final assembly and effects
Quick Start Code
from kittentts import KittenTTS
model = KittenTTS("KittenML/kitten-tts-mini-0.8")
# Generate voiceover for pipeline
model.generate_to_file("Your script text here.", "output.wav", voice="Jasper")
print("Voiceover ready for next pipeline step!")