Agora AI Noise Suppression vs DPDFNet
Agora AI Noise Suppression
agora.ioDPDFNet
github.comSide by side
| Figure | Agora AI Noise Suppression | DPDFNet |
|---|---|---|
| Price from | $0.59/mo | Free |
| Free plan | — | Yes |
| Free trial | — | — |
| Open source | — | — |
| Web | Yes | — |
| Windows | Yes | Yes |
| macOS | Yes | Yes |
| Linux | Yes | Yes |
| iOS | Yes | Yes |
| Android | Yes | Yes |
| API | — | Yes |
| Self-hosted | — | Yes |
| Plans published | 1 | 1 |
| Spec score · AI Speech Enhancement Software | 8.1 (#12) | 8.0 (#14) |
| Support | Agora says its experts provide round-the-clock support for thousands of customers in over 100 countries. | The repository directs users to open an issue or submit a pull request to add a DPDFNet integration to its community directory. |
Plans
In their own words
Agora AI Noise Suppression
Agora AI Noise Suppression is a real-time audio enhancement extension for reducing background noise during voice and video calls. Its deep-learning models target more than 100 noise types, such as fans, keyboard clicks, and crying babies, and it also reduces reverberation and echo in acoustically poor settings. Agora describes the processing as low-latency and low-power for calls and livestreams. Developers can add modular filtering through an API or enable the extension in the Agora Console. It accepts audio and runs on Android, iOS, Linux, macOS, web, and Windows. Agora lists use cases including work, gaming, education, the metaverse, livecasting, and media and entertainment. The AI Noise Suppression plan costs $0.59 a month, billed at $0.59 per 1,000 minutes; the first 10,000 minutes are free every month. Usage-based billing and customized pricing are also available. Agora says it may collect IP addresses and operational information needed to provide services, while customers manage other end-user data in their applications.
DPDFNet
DPDFNet is a free, open-source model for reducing noise in single-channel speech from recordings, live streams, and edge-device applications. It supports offline processing and streaming, and extends DeepFilterNet2 with dual-path encoder blocks and an always-on fine-tuning approach intended to reduce over-attenuation. Pretrained variants support 8, 16, and 48 kHz audio. The repository includes PyTorch code and ONNX and TFLite inference models, with command-line, Python, and stateful streaming interfaces. Streaming buffers about 20 ms before the first enhanced output, then adds about 10 ms of delay for later blocks. A hosted Gradio demo and audio examples are linked from the project page. Listed community integrations include sherpa-onnx, OBS Studio, VST3 and LADSPA plugins, Core ML ports, and WebAssembly. The project page reports real-time DPDFNet-4 deployment on Ceva-NeuPro-Nano edge NPUs in a benchmark setup using int8 weights and int16 activations. The software is under Apache License 2.0. Model files are not bundled in the source repository; it provides commands to download checkpoints and ONNX or TFLite models from Hugging Face.