Hugging Face's logo

Spaces:
thecollabagepatch
/
magenta-retry
Running

App Files Community
1
thecollabagepatch commited on
Commit
46ebda5
·
1 Parent(s): 5b3cd33

more declauding

Browse files
Files changed (1) hide show
  1. documentation.html +10 -10
documentation.html CHANGED
@@ -99,8 +99,8 @@
99
  </head>
100
  <body>
101
  <div class="header">
102
- <h1>MagentaRT Research API</h1>
103
- <p class="muted"><strong>AI Music Generation API</strong> • real-time streaming with http/ws • custom fine-tune model-switching support</p>
104
  <span class="badge">research project</span>
105
  </div>
106
 
@@ -205,12 +205,12 @@
205
 
206
  <div class="section">
207
  <h2>overview</h2>
208
- <p>this API revolves around google's magentaRT, designed for real-time audio streaming using finetunes hosted on HF. built for iOS app integration with webSocket streaming support for web applications (and potentially VSTs inside the DAW).</p>
209
  </div>
210
 
211
  <div class="section">
212
- <h2>quick start - WebSocket streaming</h2>
213
- <p>Connect to <code>wss://&lt;your-space&gt;/ws/jam</code> for real-time audio generation:</p>
214
 
215
  <h3>start real-time generation</h3>
216
  <pre><button class="copy-btn" onclick="copyCode(this)">Copy</button>{
@@ -284,14 +284,14 @@
284
 
285
  <div class="section">
286
  <h2>custom fine-tuning</h2>
287
- <p>train your own MagentaRT models and use them in the web app demo and the iOS app.</p>
288
 
289
  <div class="grid">
290
  <div class="card">
291
  <h3>1. train your model</h3>
292
  <p>use the official MagentaRT fine-tuning notebook:</p>
293
  <p><a href="https://colab.research.google.com/github/magenta/magenta-realtime/blob/main/notebooks/Magenta_RT_Finetune.ipynb" target="_blank">MagentaRT Fine-tuning Colab</a></p>
294
- <p>This will create checkpoint folders like:</p>
295
  <ul>
296
  <li><code>checkpoint_1861001/</code></li>
297
  <li><code>checkpoint_1862001/</code></li>
@@ -343,7 +343,7 @@ files.download('/content/checkpoint_1862001.tgz')</pre>
343
  </ul>
344
 
345
  <div class="info">
346
- <strong>Example Repository:</strong> <a href="https://huggingface.co/thepatch/magenta-ft" target="_blank">thepatch/magenta-ft</a><br>
347
  shows the correct file structure with .tgz files and .npy steering assets in the root directory.
348
  </div>
349
 
@@ -375,13 +375,13 @@ files.download('/content/checkpoint_1862001.tgz')</pre>
375
  </div>
376
 
377
  <div class="warning">
378
- <strong>Note:</strong> The <code>/generate_style</code> endpoint is experimental and may not properly adhere to BPM without additional context (considering metronome-based context instead of silence).
379
  </div>
380
  </div>
381
 
382
  <div class="section">
383
  <h2>deployment</h2>
384
- <p>To run your own instance:</p>
385
  <ol>
386
  <li>duplicate this huggingface space by clicking the three dots in the top right</li>
387
  <li>select 'run locally' if you got a 5090 or something</li>
 
99
  </head>
100
  <body>
101
  <div class="header">
102
+ <h1>magentaRT research API</h1>
103
+ <p class="muted"><strong>exploring ways to jam</strong> • real-time streaming with http/ws • custom fine-tune model-switching support</p>
104
  <span class="badge">research project</span>
105
  </div>
106
 
 
205
 
206
  <div class="section">
207
  <h2>overview</h2>
208
+ <p>this API revolves around google's magentaRT, designed for real-time audio streaming using finetunes hosted on HF. built for iOS app integration with webSocket streaming support for web applications (and potentially VST plugins).</p>
209
  </div>
210
 
211
  <div class="section">
212
+ <h2>quick start - webSocket streaming</h2>
213
+ <p>connect to <code>wss://&lt;your-space&gt;/ws/jam</code> for real-time audio generation:</p>
214
 
215
  <h3>start real-time generation</h3>
216
  <pre><button class="copy-btn" onclick="copyCode(this)">Copy</button>{
 
284
 
285
  <div class="section">
286
  <h2>custom fine-tuning</h2>
287
+ <p>train your own MagentaRT models and use them in the web app demo or the iOS app.</p>
288
 
289
  <div class="grid">
290
  <div class="card">
291
  <h3>1. train your model</h3>
292
  <p>use the official MagentaRT fine-tuning notebook:</p>
293
  <p><a href="https://colab.research.google.com/github/magenta/magenta-realtime/blob/main/notebooks/Magenta_RT_Finetune.ipynb" target="_blank">MagentaRT Fine-tuning Colab</a></p>
294
+ <p>this will create checkpoint folders like:</p>
295
  <ul>
296
  <li><code>checkpoint_1861001/</code></li>
297
  <li><code>checkpoint_1862001/</code></li>
 
343
  </ul>
344
 
345
  <div class="info">
346
+ <strong>example repository:</strong> <a href="https://huggingface.co/thepatch/magenta-ft" target="_blank">thepatch/magenta-ft</a><br>
347
  shows the correct file structure with .tgz files and .npy steering assets in the root directory.
348
  </div>
349
 
 
375
  </div>
376
 
377
  <div class="warning">
378
+ <strong>note:</strong> the <code>/generate_style</code> endpoint is experimental and may not properly adhere to BPM without additional context (considering metronome-based context instead of silence).
379
  </div>
380
  </div>
381
 
382
  <div class="section">
383
  <h2>deployment</h2>
384
+ <p>to run your own instance:</p>
385
  <ol>
386
  <li>duplicate this huggingface space by clicking the three dots in the top right</li>
387
  <li>select 'run locally' if you got a 5090 or something</li>