Voice: speakers and sound

TEN VAD

A lightweight speech detector for real-time voice assistants: it notices the start and end of a phrase faster than Silero VAD. Runs on servers, phones and in the browser.

Developer
Agora (TEN project), USA / China
First release
May 2025
Latest release
Jul 2025
Sizes
very small, the library is smaller than Silero VAD
License
Commercial use with conditionsApache 2.0 with additional terms: it may not be used in products competing with Agora or to let third parties build their own apps on it
Running
On your own serverAlso runs without a GPU
Industries
Customer support, Software development
Deploy on your server Hugging Face GitHub

What it does

  • Zero-lag speech detection in a voice bot
  • Fast assistant response at the end of a phrase
  • Use in mobile apps and the browser

Where it is used

Voice botsCall centersProduct development

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
no versions
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. TEN VAD 1.0 и ONNX-версия
  2. TEN VAD

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems.

Frequently asked questions

Can TEN VAD be used in a commercial project?

With conditions. License: Apache 2.0 with additional terms: it may not be used in products competing with Agora or to let third parties build their own apps on it. Restrictions vary — region, company revenue, attribution requirements. Have a lawyer check the terms before a commercial launch.

What hardware does TEN VAD need?

At minimum: Laptop or regular PC, up to 8 GB of VRAM — smaller versions. Some versions also run on an ordinary CPU, without a GPU. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does TEN VAD support Russian?

Language does not matter for this model: it does not work with text.

Where can I download TEN VAD and what does it cost?

The TEN VAD weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

Voice: speakers and soundSilero VADSilero · RussiaCommercial use allowed

The most popular open speech detector: tells voice apart from silence and noise. Processes an audio chunk in under a millisecond on a single CPU core; trained on recordings in more than 6,000 languages.

DetailsVoice: speakers and soundFireRedVADFireRedTeam (Xiaohongshu) · ChinaCommercial use allowed

A speech and sound event detector: tells apart speech, singing and music. In a 102-language test (the FLEURS set, which includes Russian) it beat Silero VAD and TEN VAD. Has a streaming mode.

DetailsVoice: speakers and soundSmart TurnDaily (Pipecat) · USACommercial use allowed

Uses intonation to tell whether a person has finished a thought or just paused, so a voice bot does not interrupt. Version 3 is 8 MB, runs on a CPU and understands 23 languages, including Russian.

Details

Source: github.com/TEN-framework/ten-vad. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.