Hugging Face Voice - Inkling

Inkling is receiving your audio directly, and replying with GGML's Qwen3TTS.

Powered by Inference Endpoints Built by Hugging Face andito A-Mahla tfrere

Tap to start

Conversation

About

This demo sends your audio directly to Inkling on Hugging Face Inference Endpoints, then voices Inkling's reply with Qwen3-TTS through GGML.

View the project on GitHub

The pipeline

  1. You speak
  2. VAD detects speech silero-vad
  3. Audio LLM hears you and composes the reply · via Inference Endpoints thinkingmachines/Inkling-NVFP4
  4. TTS speaks it back Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice
  5. The orb replies

Settings

Noise gate Off
Off −3 dB
Mutes the mic below the handle so room noise isn't sent. Slide fully left to turn it off; the bar shows your live input while in a call.
Connect first, then come back here to apply live changes.

Tools

Let the assistant act during the conversation. Changes apply live.

Web search Look things up on Google, via Serper.
Camera Let the assistant see through your webcam.
A live preview shows bottom-left while the camera is on.

That's a wrap for now