Neural audio codecs: how to get audio into LLMs

Re:Neural audio codecs: how to get audio into LLMs by MichealCodes on Tuesday October 21, 2025 @09:40AM (#45655692)

I don't think we've had the transformer moment for audio training yet, but yes, in theory audio-first models will be much more capable.

[ Parent | on HN ]

Re:Neural audio codecs: how to get audio into LLMs by trollbridge on Tuesday October 21, 2025 @09:42AM (#45655709)

Particularly interesting would be transformations between tokenised audio and tokenised text.

I recall someone telling me once up to 90% of communication can be non-verbal, so when an LLM sticks to just text, it's only getting 10% of the data.

[ Parent | on HN ]