▲ 494 ▼ VLC tops 6 billion downloads, previews AI-generated subtitles (techcrunch.com) submitted 2 years ago by neme@lemm.ee to c/opensource@programming.dev 127 comments fedilink hide all child comments
[–] themeatbridge@lemmy.world 2 points 2 years ago (1 child) That would depend on the LLM and the data used to train it. permalink fedilink source parent hideshow 2 child comments replies: [–] Aatube@kbin.melroy.org 3 points 2 years ago (1 child) IIRC you can't use LLMs for this. permalink fedilink source parent hideshow 2 child comments replies: [–] themeatbridge@lemmy.world 1 point 2 years ago (1 child) I didn't read the article, but I would have assumed that the AI was using predictive text to guess at the next word. Speech recognition is already pretty good, but it often misses contextual cues that an LLM would be good at spotting. Like, "The famous French impressionist painter mayonnaise..." permalink fedilink source parent hideshow 2 child comments replies: [–] kautau@lemmy.world 4 points 2 years ago* (last edited 2 years ago) (1 child) Probably something like https://github.com/openai/whisper which isn’t an LLM, but is a different type of model dedicated to speech recognition permalink fedilink source parent hideshow 2 child comments replies: [–] themeatbridge@lemmy.world 1 point 2 years ago That makes sense. permalink fedilink source parent
[–] Aatube@kbin.melroy.org 3 points 2 years ago (1 child) IIRC you can't use LLMs for this. permalink fedilink source parent hideshow 2 child comments replies: [–] themeatbridge@lemmy.world 1 point 2 years ago (1 child) I didn't read the article, but I would have assumed that the AI was using predictive text to guess at the next word. Speech recognition is already pretty good, but it often misses contextual cues that an LLM would be good at spotting. Like, "The famous French impressionist painter mayonnaise..." permalink fedilink source parent hideshow 2 child comments replies: [–] kautau@lemmy.world 4 points 2 years ago* (last edited 2 years ago) (1 child) Probably something like https://github.com/openai/whisper which isn’t an LLM, but is a different type of model dedicated to speech recognition permalink fedilink source parent hideshow 2 child comments replies: [–] themeatbridge@lemmy.world 1 point 2 years ago That makes sense. permalink fedilink source parent
[–] themeatbridge@lemmy.world 1 point 2 years ago (1 child) I didn't read the article, but I would have assumed that the AI was using predictive text to guess at the next word. Speech recognition is already pretty good, but it often misses contextual cues that an LLM would be good at spotting. Like, "The famous French impressionist painter mayonnaise..." permalink fedilink source parent hideshow 2 child comments replies: [–] kautau@lemmy.world 4 points 2 years ago* (last edited 2 years ago) (1 child) Probably something like https://github.com/openai/whisper which isn’t an LLM, but is a different type of model dedicated to speech recognition permalink fedilink source parent hideshow 2 child comments replies: [–] themeatbridge@lemmy.world 1 point 2 years ago That makes sense. permalink fedilink source parent
[–] kautau@lemmy.world 4 points 2 years ago* (last edited 2 years ago) (1 child) Probably something like https://github.com/openai/whisper which isn’t an LLM, but is a different type of model dedicated to speech recognition permalink fedilink source parent hideshow 2 child comments replies: [–] themeatbridge@lemmy.world 1 point 2 years ago That makes sense. permalink fedilink source parent