Multimodal Models
Enabling models to natively ingest images and audio via a projection model
Speech to Text
Transcribe spoken audio to text with NobodyWho in Python.
Text to Speech
Generate WAV audio from text with NobodyWho in Python.
Version: main
Vision & Audio
Multimodal Models cover LLMs that can natively consume images and audio directly, without a separate transcription step. For converting spoken audio into text, see Speech-to-Text. For generating natural-sounding voice from text, see Text-to-Speech.