Hacker News: bazlanhttps://news.ycombinator.com/user?id=bazlanHacker News RSShttps://hnrss.org/hnrss v2.1.1Wed, 29 Apr 2026 20:29:56 +0000<![CDATA[New comment by bazlan in "VibeVoice: A Frontier Open-Source Text-to-Speech Model"]]>Sad to not see vui on the comparisons!

A 100M podcast model

https://huggingface.co/spaces/fluxions/vui-space

]]>Wed, 03 Sep 2025 18:03:40 +0000https://news.ycombinator.com/item?id=45118740bazlanhttps://news.ycombinator.com/item?id=45118740https://news.ycombinator.com/item?id=45118740<![CDATA[Open Source NotebookLM – 100M]]>Article URL: https://twitter.com/harrycblum/status/1930709683242713496

Comments URL: https://news.ycombinator.com/item?id=44216584

Points: 18

# Comments: 5

]]>Sun, 08 Jun 2025 12:28:16 +0000https://twitter.com/harrycblum/status/1930709683242713496bazlanhttps://news.ycombinator.com/item?id=44216584https://news.ycombinator.com/item?id=44216584<![CDATA[New comment by bazlan in "Show HN: Dia, an open-weights TTS model for generating realistic dialogue"]]>fluxions.ai has a similar model

]]>Tue, 22 Apr 2025 09:00:42 +0000https://news.ycombinator.com/item?id=43760254bazlanhttps://news.ycombinator.com/item?id=43760254https://news.ycombinator.com/item?id=43760254<![CDATA[New comment by bazlan in "Launch HN: Hamming (YC S24) – Automated Testing for Voice Agents"]]>As someone who has worked in TTS for over 4 years now. I can tell you that evaluation is the most difficult aspect of generative audio ML.

How will this really check that the models are performing well vs just listening?

]]>Fri, 16 Aug 2024 06:15:30 +0000https://news.ycombinator.com/item?id=41263610bazlanhttps://news.ycombinator.com/item?id=41263610https://news.ycombinator.com/item?id=41263610<![CDATA[New comment by bazlan in "PHP Doesn't Suck Anymore"]]>Let it die

]]>Sat, 04 May 2024 10:03:36 +0000https://news.ycombinator.com/item?id=40256344bazlanhttps://news.ycombinator.com/item?id=40256344https://news.ycombinator.com/item?id=40256344