TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

End-to-End Amp Modeling: From Data to Controllable Guitar Amplifier Models

2 点作者 _greim_10 个月前

2 条评论

_greim_10 个月前
As I understand, this company&#x27;s existing consumer-level product can already be used to capture the sound of an amplifier, but it&#x27;s a snapshot of the device&#x27;s (amplifier, effects pedal) settings in terms gain, eq, etc.<p>Meanwhile, they have an <i>extremely</i> labor-intensive set of techniques for modeling a device&#x27;s analog circuitry, resulting in a model that allows the user to adjust gain, eq, etc. This isn&#x27;t a consumer-level process; it happens in a laboratory somewhere, and the output is shipped as a software plugin or model on a digital effects unit.<p>This technology bridges the gap. Ultimately it&#x27;s an unguided ML approach akin to the former, but introduces ML-guided robotic knob-turning (AKA &quot;TINA&quot;) which (unlike the former) maps continuous changes within the device&#x27;s parameter space, allowing to ship something more like the latter.
_greim_10 个月前
Accompanying public announcements:<p><a href="https:&#x2F;&#x2F;neuraldsp.com&#x2F;quad-cortex-updates&#x2F;introducing-tina" rel="nofollow">https:&#x2F;&#x2F;neuraldsp.com&#x2F;quad-cortex-updates&#x2F;introducing-tina</a><p><a href="https:&#x2F;&#x2F;neuraldsp.com&#x2F;news&#x2F;neural-dsp-amplifier-modeling-technology" rel="nofollow">https:&#x2F;&#x2F;neuraldsp.com&#x2F;news&#x2F;neural-dsp-amplifier-modeling-tec...</a>