#tts

Horus.F5Tts.Onnx Here's what landed in v0.3.0: Streaming: hear the first sentence...

Free MIT-licensed desktop app for voice cloning, dictation, and TTS. What the 3-second seed claim leaves out — and how to actually run it.

The landscape for open-weight text-to-speech (TTS) has fundamentally shifted. We have reached a point...

What it took to package Miso TTS 8B for local NVIDIA GPUs, Windows launchers, Blackwell cards, and a two-speed voice demo.

How we got sad / happy / angry / fearful speech to work on cloned voices in a pure-C Qwen3-TTS engine — a 25 MB 'graft' clone that keeps the emotion levers alive, a steering + fine-tune recipe hard-won after many dead ends, and — because emotion is a vector — blended 'dyad' emotions plus switching emotion mid-sentence from a single prompt.

You spent hours scripting the perfect episode. It's got a cold open, three interview-style segments,...

Most people meet an AI voice and immediately reach for the controls. They tweak speed, hunt for the...

Running an open-source AI short-video generator end-to-end on a local stack — topic in, vertical MP4 out — with a local Ollama model and no paid LLM keys.

I took NVIDIA's pdf-to-podcast blueprint and ran it on a fully local stack — tiered Claude writes the script, Kokoro does the voices, no NVIDIA or ElevenLabs keys.

You spent weeks preparing that webinar. You showed up live, fielded questions, and delivered real...

Originally published on andrew.ooo — visit the original for any updates, code snippets that aged...

I benchmarked local voice-cloning models across English, German, Modern Standard Arabic, Spanish, and...