{"author":"mrkn1","children":[{"author":"ig-88ms","children":[{"author":"mrkn1","children":[],"created_at":"2026-11-18T16:53:06.000Z","created_at_i":1795020786,"id":50548646,"options":[],"parent_id":50548405,"points":null,"story_id":50547122,"text":"i dont like it either, but you can train your own voice with 10h of audio.","title":null,"type":"comment","url":null},{"author":"chandradot99","children":[],"created_at":"2026-11-19T07:44:49.000Z","created_at_i":1795074289,"id":50557169,"options":[],"parent_id":50548405,"points":null,"story_id":50547122,"text":"yeah pretty much same for me, here are other small models which you can try kitten tts(about 7m), kokoro(82m but still runs on cpu).","title":null,"type":"comment","url":null}],"created_at":"2026-11-18T16:35:17.000Z","created_at_i":1795019717,"id":50548405,"options":[],"parent_id":50547122,"points":null,"story_id":50547122,"text":"I was irritated by the childlike voice at first, but they acknowledge that they were aiming for a twenty something female voice and plan to build a more mature voice.","title":null,"type":"comment","url":null},{"author":"ilyaizen","children":[],"created_at":"2026-11-19T00:13:59.000Z","created_at_i":1795047239,"id":50554233,"options":[],"parent_id":50547122,"points":null,"story_id":50547122,"text":"Hi there.<p>I made CopySpeech, a desktop app that reads copied text aloud. Copy the same text twice to hear it.<p>The default voice is tiny-voca, a 4.5M-parameter model. In my tests, it starts speaking about 0.1 seconds after a warm start. I&#x27;m happy with how fast it is.<p>Repo: <a href=\"https:&#x2F;&#x2F;github.com&#x2F;ilyaizen&#x2F;CopySpeech\" rel=\"nofollow\">https:&#x2F;&#x2F;github.com&#x2F;ilyaizen&#x2F;CopySpeech</a>\nSite: <a href=\"https:&#x2F;&#x2F;copyspeech.ilyaizen.link&#x2F;\" rel=\"nofollow\">https:&#x2F;&#x2F;copyspeech.ilyaizen.link&#x2F;</a>","title":null,"type":"comment","url":null}],"created_at":"2026-11-18T15:01:14.000Z","created_at_i":1795014074,"id":50547122,"options":[],"parent_id":null,"points":43,"story_id":50547122,"text":null,"title":"A 4.5M-parameter TTS model runs 70x real-time on CPU and cost just $10 to train","type":"story","url":"https://huggingface.co/n4ze3m/tiny-voca"}
