https://hackerpaper.kaipfister.de/p/1608.01281/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=43901785urn:hackerpapers:mention:1608.01281:43901785Tue, 06 May 2025 04:22:23 GMT<p>Think of it as generating a constantly streaming infinite list of latents. These latents are basically decoded to a tuple [time_until_my_turn(latent_t), audio(latent_t)]. You can train it to minimize the error of its time_until_my_turn predictions from ground truth of training samples, as well as the quality of the audio generated. Basically a change-point prediction model. Ilya Sutskever (among others) worked on something like this long ago, it might have inspired OpenAI&#x27;s advanced voice models: &gt; Sequence-to-sequence models with soft attention had significant success in machine translation…</p><p><a href="https://hackerpaper.kaipfister.de/p/1608.01281/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.LGhttps://news.ycombinator.com/item?id=12301476urn:hackerpapers:submission:1608.01281:12301476Tue, 16 Aug 2016 23:40:25 GMT<p>Learning Online Alignments with Continuous Rewards Policy Gradient</p><p></p><p>2 points; 0 discussion replies on HN. <a href="https://hackerpaper.kaipfister.de/p/1608.01281/">Paper and source history</a>. A submission is not an endorsement.</p>cs.LG