{"channel":"cities","content":"Huggingface had approximately one open-source model that supported Lithuanian TTS.\r\n\r\nFortunately, one model is all that is needed.\r\n\r\n----\r\n\r\nIt is called \"OuteTTS\" ( Huggingface https://huggingface.co/OuteAI/Llama-OuteTTS-1.0-1B ; website https://outeai.com ).\r\n\r\nAnd, after a bit of handholding ( and a sample audio file from OpenAI.fm to generate a \"speaker profile\"), it works. All the letters are pronounced.\r\n\r\n----\r\n\r\nIt is a local model. Which means that it will take a few hours to generate ( checks notes) 723 words/phrases to read. ( well, actually, 2169 if I keep doing three different voices for it)","created_at":"2025-05-30T03:23:16.802999","id":566,"llm_annotations":{},"parent_id":564,"processed_content":"

Huggingface had approximately one open-source model that supported Lithuanian TTS.\r

\n

Fortunately, one model is all that is needed.\r

\n
\n

It is called \"OuteTTS\" \u2699\ufe0f Huggingface https://huggingface.co/OuteAI/Llama-OuteTTS-1.0-1B ; website https://outeai.com .\r

\n

And, after a bit of handholding \ud83d\udca1 and a sample audio file from OpenAI.fm to generate a \"speaker profile\", it works. All the letters are pronounced.\r

\n
\n

It is a local model. Which means that it will take a few hours to generate \ud83d\udd25 checks notes 723 words/phrases to read. \u2694\ufe0f well, actually, 2169 if I keep doing three different voices for it

","quotes":[],"subject":"faulkton (part 2)"}