Auphonic vs DPDFNet

Auphonic is an automated audio post-production web service for podcasts, broadcasts, radio, interviews, audiobooks, lectures and other audio or video projects. Its algorithms can reduce noise and reverb, adjust levels, filter and enhance recordings, meet loudness specifications, and remove filler words, coughs and silence. Multitrack processing creates mixdowns with automatic ducking, noise gates and mic-bleed removal. Auphonic also provides multilingual speech-to-text, automatic shownotes and chapters, and a shareable transcript editor. It supports common audio and video formats, waveform audiograms, metadata, chapters and burned-in styled subtitles. Integrations include storage, publishing, automation and speech-recognition services; workflows can use the API, CLI, watch folders and batch productions. The free plan is free and includes two hours per month, but free accounts do not include multilingual speech recognition, automatic shownotes and chapters, watch folders or batch productions. Free multitrack productions must be under 20 minutes. Billing has a three-minute minimum per production, and productions are deleted after 21 days for audio and API work or seven days for video.

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.