Audio LLM

Unlike Large Model (LLM) nodes in GPTBots, the Audio LLM supports both audio input and audio output. What's more, it allows users to specify the applied voice tone.

Node Configuration

• Large Language Model (LLM): Same as the configuarion in Large Model (LLM) node, but only an audio LLM can be selected.

• Audio Input: Only variables of the "file-audio" type are supported.

• Voice Tone: Defines the tone of the output audio.

• For descriptions of the other configuration items (AI Creativity, Maximum Response, Identity Prompt, User Prompt, Memory, Tool, and Data Table, please refer to Large Model.

Node Output

Outputs of this node only include the following three fields:

• text: Text content generated by the LLM.

• audio: Audio file generated by the LLM, provided as a URL.

• transcript: Transcription text of the audio file.