Download index.html from danielrosehill/Audio-To-Audio-Index: direct link, hf CLI and curl.
- Browser
- Download file 12.8 kB
-
https://huggingface.co/spaces/danielrosehill/Audio-To-Audio-Index/resolve/main/index.html
- Command line
-
hf download hf://spaces/danielrosehill/Audio-To-Audio-Index/index.html
-
curl -L -o index.html https://huggingface.co/spaces/danielrosehill/Audio-To-Audio-Index/resolve/main/index.html
12.8 kB
| <html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width,initial-scale=1"> | |
| <title>Audio-to-Audio Index</title><style> | |
| :root{--bg:#f7f7f4;--card:#fff;--ink:#1d2327;--muted:#5d676e;--line:#dfe2dd;--accent:#0b6e69;--soft:#e3f1ef} | |
| *{box-sizing:border-box}body{margin:0;background:var(--bg);color:var(--ink);font:16px/1.5 system-ui,-apple-system,"Segoe UI",sans-serif} | |
| main{max-width:1080px;margin:0 auto;padding:32px 20px 56px}a{color:var(--accent)}h1{font-size:2rem;margin:0 0 8px} | |
| h2{font-size:1.15rem;margin:34px 0 4px}.lede,.about{color:var(--muted);max-width:72ch;margin:0 0 12px} | |
| nav.toc{display:flex;flex-wrap:wrap;gap:8px;margin:14px 0 18px}nav.toc a{background:var(--soft);padding:5px 11px;border-radius:999px;text-decoration:none;font-size:.9rem} | |
| nav.toc span{color:var(--muted)}input{width:100%;padding:9px 12px;border:1px solid var(--line);border-radius:6px;font:inherit} | |
| .grid{display:grid;grid-template-columns:repeat(auto-fill,minmax(300px,1fr));gap:12px;margin-top:10px} | |
| .card{display:block;background:var(--card);border:1px solid var(--line);border-radius:8px;padding:12px 14px;text-decoration:none;color:var(--ink)} | |
| .card:hover{border-color:var(--accent)}.card b{display:block;word-break:break-word}.card small{color:var(--muted)}.card p{margin:6px 0 0;font-size:.92rem} | |
| .kind{float:right;font-size:.72rem;text-transform:uppercase;letter-spacing:.04em;color:var(--muted)}.kind.space{color:#7a4fb0} | |
| footer{margin-top:40px;color:var(--muted);font-size:.85rem}</style></head><body><main> | |
| <h1>Audio-to-Audio Index</h1><p class="lede">Speech enhancement, restoration and audio editing models on the Hugging Face Hub, grouped by what they do. Curated by Daniel Rosehill; source collections: Voice Enhancement and Audio editing utilities.</p><p class="lede">27 entries · Collections: <a href="https://huggingface.co/collections/danielrosehill/voice-enhancement-68faa39c1ebef0f22b7faaad">Voice Enhancement</a> · <a href="https://huggingface.co/collections/danielrosehill/audio-editing-utilities-68f3f55204fbae5be954546e">Audio editing utilities</a></p> | |
| <input type="search" id="q" placeholder="Filter by name or description"><nav class="toc"><a href="#g0">Universal enhancement and restoration <span>9</span></a><a href="#g1">Dialogue and speaker-aware <span>2</span></a><a href="#g2">Real-time and on-device denoising <span>7</span></a><a href="#g3">Bandwidth extension and upscaling <span>3</span></a><a href="#g4">Ready-to-run builds <span>2</span></a><a href="#g5">Instruction-guided audio editing <span>2</span></a><a href="#g6">Try in the browser and reference <span>2</span></a></nav> | |
| <section id="g0"><h2>Universal enhancement and restoration</h2><p class="about">One model for noise, reverb, codec damage and band-limiting. Mostly generative; best quality, not real-time.</p><div class="grid"><a class="card" href="https://huggingface.co/nvidia/NVIDIA-NemotronLabs-RE-USE" data-q="nvidia/nvidia-nemotronlabs-re-use nvidia's current re-use release: universal speech enhancement (mamba), any sampling rate, language-agnostic."><span class="kind model">model</span><b>NVIDIA-NemotronLabs-RE-USE</b><small>nvidia</small><p>NVIDIA's current RE-USE release: universal speech enhancement (Mamba), any sampling rate, language-agnostic.</p></a><a class="card" href="https://huggingface.co/nvidia/RE-USE" data-q="nvidia/re-use the original re-use release."><span class="kind model">model</span><b>RE-USE</b><small>nvidia</small><p>The original RE-USE release.</p></a><a class="card" href="https://huggingface.co/cisco-ai/stupase" data-q="cisco-ai/stupase stupase: generative enhancement aiming at studio quality; removes noise and reverb, keeps speaker identity. 16 khz."><span class="kind model">model</span><b>stupase</b><small>cisco-ai</small><p>StuPASE: generative enhancement aiming at studio quality; removes noise and reverb, keeps speaker identity. 16 kHz.</p></a><a class="card" href="https://huggingface.co/sarulab-speech/sidon-v0.1" data-q="sarulab-speech/sidon-v0.1 sidon: speech restoration to clean studio-like audio."><span class="kind model">model</span><b>sidon-v0.1</b><small>sarulab-speech</small><p>Sidon: speech restoration to clean studio-like audio.</p></a><a class="card" href="https://huggingface.co/jadechoghari/VoiceRestore" data-q="jadechoghari/voicerestore voicerestore: flow-matching transformer for degraded voice recordings."><span class="kind model">model</span><b>VoiceRestore</b><small>jadechoghari</small><p>VoiceRestore: flow-matching transformer for degraded voice recordings.</p></a><a class="card" href="https://huggingface.co/ResembleAI/resemble-enhance" data-q="resembleai/resemble-enhance resemble enhance: denoiser plus generative enhancer; widely used."><span class="kind model">model</span><b>resemble-enhance</b><small>ResembleAI</small><p>Resemble Enhance: denoiser plus generative enhancer; widely used.</p></a><a class="card" href="https://huggingface.co/nineninesix/diamond-1.0" data-q="nineninesix/diamond-1.0 codec-based autoregressive speech restoration."><span class="kind model">model</span><b>diamond-1.0</b><small>nineninesix</small><p>Codec-based autoregressive speech restoration.</p></a><a class="card" href="https://huggingface.co/disco-eth/DAC-SE1" data-q="disco-eth/dac-se1 dac-se1: enhancement as sequence modelling over discrete audio tokens."><span class="kind model">model</span><b>DAC-SE1</b><small>disco-eth</small><p>DAC-SE1: enhancement as sequence modelling over discrete audio tokens.</p></a><a class="card" href="https://huggingface.co/sp-uhh/speech-enhancement-sgmse" data-q="sp-uhh/speech-enhancement-sgmse sgmse+: diffusion-based enhancement and dereverberation; research reference."><span class="kind model">model</span><b>speech-enhancement-sgmse</b><small>sp-uhh</small><p>SGMSE+: diffusion-based enhancement and dereverberation; research reference.</p></a></div></section><section id="g1"><h2>Dialogue and speaker-aware</h2><p class="about">Clean speech while separating or suppressing other speakers.</p><div class="grid"><a class="card" href="https://huggingface.co/sarulab-speech/DialogueSidon2" data-q="sarulab-speech/dialoguesidon2 two-speaker dialogue: separates and restores each speaker to a clean 24 khz track."><span class="kind model">model</span><b>DialogueSidon2</b><small>sarulab-speech</small><p>Two-speaker dialogue: separates and restores each speaker to a clean 24 kHz track.</p></a><a class="card" href="https://huggingface.co/weya-ai/hush" data-q="weya-ai/hush real-time denoise that also suppresses background speakers; onnx."><span class="kind model">model</span><b>hush</b><small>weya-ai</small><p>Real-time denoise that also suppresses background speakers; ONNX.</p></a></div></section><section id="g2"><h2>Real-time and on-device denoising</h2><p class="about">Small or streaming models for calls, live audio and phones.</p><div class="grid"><a class="card" href="https://huggingface.co/nvidia/Real-time_RE-USE" data-q="nvidia/real-time_re-use streaming re-use."><span class="kind model">model</span><b>Real-time_RE-USE</b><small>nvidia</small><p>Streaming RE-USE.</p></a><a class="card" href="https://huggingface.co/desert-ant-labs/clear" data-q="desert-ant-labs/clear lightweight denoise and dereverb; onnx and tflite builds."><span class="kind model">model</span><b>clear</b><small>desert-ant-labs</small><p>Lightweight denoise and dereverb; ONNX and TFLite builds.</p></a><a class="card" href="https://huggingface.co/LocalAI-io/LocalVQE" data-q="localai-io/localvqe noise suppression plus acoustic echo cancellation; gguf."><span class="kind model">model</span><b>LocalVQE</b><small>LocalAI-io</small><p>Noise suppression plus acoustic echo cancellation; GGUF.</p></a><a class="card" href="https://huggingface.co/soniqo/DeepFilterNet3-ONNX" data-q="soniqo/deepfilternet3-onnx deepfilternet3 as onnx (upstream lives on github)."><span class="kind model">model</span><b>DeepFilterNet3-ONNX</b><small>soniqo</small><p>DeepFilterNet3 as ONNX (upstream lives on GitHub).</p></a><a class="card" href="https://huggingface.co/JacobLinCool/MP-SENet-DNS" data-q="jacoblincool/mp-senet-dns mp-senet trained on dns challenge data."><span class="kind model">model</span><b>MP-SENet-DNS</b><small>JacobLinCool</small><p>MP-SENet trained on DNS Challenge data.</p></a><a class="card" href="https://huggingface.co/starkdmi/MossFormer2-SE" data-q="starkdmi/mossformer2-se mossformer2 enhancement, mlx port for apple silicon."><span class="kind model">model</span><b>MossFormer2-SE</b><small>starkdmi</small><p>MossFormer2 enhancement, MLX port for Apple silicon.</p></a><a class="card" href="https://huggingface.co/speechbrain/metricgan-plus-voicebank" data-q="speechbrain/metricgan-plus-voicebank metricgan+ via speechbrain; classic baseline."><span class="kind model">model</span><b>metricgan-plus-voicebank</b><small>speechbrain</small><p>MetricGAN+ via SpeechBrain; classic baseline.</p></a></div></section><section id="g3"><h2>Bandwidth extension and upscaling</h2><p class="about">Recover high frequencies from low-rate or compressed audio.</p><div class="grid"><a class="card" href="https://huggingface.co/audiojs/restore" data-q="audiojs/restore denoise, dereverb and bandwidth extension in onnx (en/fr)."><span class="kind model">model</span><b>restore</b><small>audiojs</small><p>Denoise, dereverb and bandwidth extension in ONNX (EN/FR).</p></a><a class="card" href="https://huggingface.co/spaces/sereich/BroadcastAudioUpscaling" data-q="sereich/broadcastaudioupscaling space: upscale broadcast-quality audio."><span class="kind space">space</span><b>BroadcastAudioUpscaling</b><small>sereich</small><p>Space: upscale broadcast-quality audio.</p></a><a class="card" href="https://huggingface.co/spaces/patriotyk/Apollo" data-q="patriotyk/apollo space: apollo restoration of lossy-compressed audio."><span class="kind space">space</span><b>Apollo</b><small>patriotyk</small><p>Space: Apollo restoration of lossy-compressed audio.</p></a></div></section><section id="g4"><h2>Ready-to-run builds</h2><p class="about">GGUF conversions for CPU runtimes.</p><div class="grid"><a class="card" href="https://huggingface.co/cstr/Sidon-GGUF" data-q="cstr/sidon-gguf sidon in gguf."><span class="kind model">model</span><b>Sidon-GGUF</b><small>cstr</small><p>Sidon in GGUF.</p></a><a class="card" href="https://huggingface.co/audio-cpp/RE-USE-GGUF" data-q="audio-cpp/re-use-gguf re-use in gguf."><span class="kind model">model</span><b>RE-USE-GGUF</b><small>audio-cpp</small><p>RE-USE in GGUF.</p></a></div></section><section id="g5"><h2>Instruction-guided audio editing</h2><p class="about">Edit audio from a text prompt: add, remove or replace sounds.</p><div class="grid"><a class="card" href="https://huggingface.co/GeWu-Lab/SteerSound" data-q="gewu-lab/steersound add, remove or replace sounds and change pitch or volume from a prompt (flow-matching dit)."><span class="kind model">model</span><b>SteerSound</b><small>GeWu-Lab</small><p>Add, remove or replace sounds and change pitch or volume from a prompt (flow-matching DiT).</p></a><a class="card" href="https://huggingface.co/jac22/UNISON" data-q="jac22/unison unified sound generation and editing: text-to-audio, zero-shot tts, speech editing. en/zh, non-commercial licence."><span class="kind model">model</span><b>UNISON</b><small>jac22</small><p>Unified sound generation and editing: text-to-audio, zero-shot TTS, speech editing. EN/ZH, non-commercial licence.</p></a></div></section><section id="g6"><h2>Try in the browser and reference</h2><div class="grid"><a class="card" href="https://huggingface.co/spaces/drewThomasson/DeepFilterNet2_no_limit" data-q="drewthomasson/deepfilternet2_no_limit deepfilternet2 demo with no length limit."><span class="kind space">space</span><b>DeepFilterNet2_no_limit</b><small>drewThomasson</small><p>DeepFilterNet2 demo with no length limit.</p></a><a class="card" href="https://huggingface.co/spaces/eustlb/audio-to-audio-models-on-the-hub" data-q="eustlb/audio-to-audio-models-on-the-hub how many audio-to-audio models are on the hub."><span class="kind space">space</span><b>audio-to-audio-models-on-the-hub</b><small>eustlb</small><p>How many audio-to-audio models are on the Hub.</p></a></div></section> | |
| <footer>Links only: models are their authors' work, under their own licences. Maintained in danielrosehill/HF-Collections.</footer> | |
| </main><script> | |
| const q=document.getElementById('q');q.addEventListener('input',()=>{const v=q.value.toLowerCase().trim(); | |
| document.querySelectorAll('.card').forEach(c=>c.style.display=!v||c.dataset.q.includes(v)?'':'none'); | |
| document.querySelectorAll('section').forEach(s=>s.style.display=[...s.querySelectorAll('.card')].some(c=>c.style.display!=='none')?'':'none');}); | |
| </script></body></html> |