Hey, I’m glad you’re still here. My name’s Fed.
I’m a Voice AI specialist with nearly two decades in the field: ASR, TTS, conversation design, speaker and language identification. My focus sits at the intersection of Philippine languages and low-resource language technology, which is a fancy way of saying I care deeply about building voice tech for languages the industry often overlooks.
If there’s a thread through my work, it’s this: I don’t trust numbers without context. I’ve spent years watching speech AI get misjudged by people who see a benchmark score and assume they understand what it means. They don’t see the normalization decisions, the domain mismatches, the fundamental differences in how audio signals behave compared to text or images. Voice AI launched the neural network era, yet it remains the least understood modality, and the most unfairly compared. I have feelings about this, as you can probably tell.
Beyond the technical work, I have a way of moving through the world. Anything goes, but nothing gets a free pass. I believe in thinking from first principles and discarding inherited baggage that doesn’t survive scrutiny. If an idea is truly essential, it should be able to stand up to the question: “Why this, and not something else?” I try to stay open-minded, just not so open that the floor falls out.
I’m based in Japan, originally from the Philippines. If you’re looking for my formal portfolio and professional background, that lives at angfederi.co. But if something here resonates, or even if you just want to say hello, I’d love to hear from you.