ASR and Speech Translation demo powered by FAMA models
Determine GPU requirements for large language models