Spoken replies to your speech
You speak, and Personaplex answers with audio in WAV at 24kHz.
Transform audio
PersonaPlex is a real-time, full-duplex speech-to-speech conversational model that enables persona control through text-based role prompts and audio-based voice conditioning.
Opens the audio studio with Personaplex ready to use. You can look around as a guest. Running it takes the Pro plan or higher.
What it makes
Transform audio
Changes an audio clip into a new one.
What you need
Optional: a sample of the voice to copy.
What you get
A new audio clip
Personaplex answers a clip of your speech with spoken audio. You add your clip, describe the AI persona in a text Prompt, and get the spoken reply back.
Audio, your speech, is the only required input. Prompt describes the persona and the conversation, for example a wise and friendly teacher. Voice has ten presets, labelled natural or variety and female or male. Voice Audio is a voice sample, and 10 seconds or more of clear speech is recommended. When you add it, the reply uses that cloned voice and Voice is ignored. Temperature Audio and Temperature Text each go from 0 to 2, and higher values produce more diverse outputs.
The reply is WAV audio at 24 kilohertz (kHz). Personaplex costs 100 credits, and a run that uses Voice Audio is billed at twice the rate.
You speak, and Personaplex answers with audio in WAV at 24kHz.
Describe who the AI is and what the conversation is about in plain text.
Pick one of ten preset voices, or add a Voice Audio clip so the reply uses a cloned voice.
Add the audio clip of your speech.
Write the Prompt, then pick a Voice or add a Voice Audio clip.
Start the run and play the spoken reply.
Write the persona in the Prompt, speak to it, and hear its answer aloud.
Record your question and get the reply back as audio.
Quick answers to help you get the most out of Personaplex in FunAI Studio.