Inflect-Micro-v2: Complete Local Text-to-Waveform TTS Under 10M Parameters
Summary
Inflect-Micro-v2 is a compact local TTS model designed for offline use with under 10M parameters. It provides CPU/CUDA inference, a fixed English voice, and an Apache-2.0 license, with run-locally instructions and ONNX export paths. The article highlights its footprint, speed, and evaluative metrics to compare against Nano and other compact TTS models, making it relevant for SMB IT and AI tools coverage.