Source Intelligence
人声清晰度。 创作自由。
什么是 Source Intelligence？
Source Intelligence 是一项实时人声增强技术，利用独特的声源分离算法，将人声从背景噪音中分离出来。它运行在值得信赖的 L-ISA Processor II Core 上，可消除人群噪音、乐器串音、PA 溢出和房间混响。
Source Intelligence 无需使用激进的均衡器、噪声门、阈值或反应式处理，即可提供更清晰、更易理解的人声。这显著提高了反馈前增益，并带来水晶般清晰的人声混音，同时不失表演的情感冲击力和亲密感。
凭借更出色的人声隔离和无与伦比的音频质量，工程师获得了更富创意地进行混音的自由，而表演者和演讲者则可以自由地在舞台上的任何位置移动——甚至在 PA 前方——而无需牺牲清晰度、存在感或担心反馈。
应用
重新定义现场人声清晰度。
品质
一个强大的混音始于清晰、可控的人声。Source Intelligence 在麦克风处减少噪音、串音和不想要的舞台声音，从一开始就提供干净的输入信号。由于需要的校正处理更少，工程师可以花更少的时间进行修复，而有更多的时间混音。
可靠性
Source Intelligence 运行在 L-ISA Processor II Core 上——这是一个为创意性而又要求严苛的巡回演出或固定安装环境而设计的平台。其性能稳定、可靠，并通过数百小时的连续测试以及与顶尖 FOH 工程师在真实世界部署中的广泛验证。
简洁性
Source Intelligence 专为现场扩声的现实而设计：快速设置、直观的控制和极少的调校。没有阈值、比率或复杂的菜单需要调整——只有直接了当的通道路由和设置。其结果是，通过更简单、更高效的工作流程，获得一致的高质量输出。
谁将从 Source Intelligence 中受益？
消除舞台噪音、人群噪音和 PA 溢出，即使是如耳语一般的人声表演也能清晰可闻，并具有情感冲击力。
消除 PA 串音，让表演者无拘无束地表达自己。麦克风反馈已成为过去时——即使在将表演者置于 PA 前方的创新舞台设置上表演时也是如此。
Source Intelligence 从人声中移除所有不想要的声音，让工程师能够专注于人声混音的创意，而不是校正处理。
Source Intelligence 赋予音响和舞台设计师无限的创新自由。将表演者放置在舞台上的任何位置，而无需担心反馈或舞台噪音。
应用
巡回演出
“Source Intelligence 让我获得了前所未有的人声效果。这基本上是录音室级别的人声，完全改变了游戏规则……不要等待。因为使用它的工程师将永远比你更有优势。”
- Joey Diehl（Benson Boone 的 FOH 工程师）
音乐剧
“它让我安心，知道演员的麦克风不会失控。我知道我可以让麦克风保持开启状态，而通常情况下，我会对演员身上的全指向麦克风更加小心。”
- Cody Spencer（《Dog Day Afternoon》音响设计师）
宗教场所
“[Source Intelligence] 让我们能够从麦克风中获得我们想要的增益，而无需担心反馈或其他可能在你将人声置于理想位置时影响混音的舞台噪音。”
- Luke Roetman (Applications Director, North Point Ministries)
The Magic Behind Source Intelligence
Intuitive Controls
- Four instances.
- Independent controls for refinement of high and low frequency content.
- Input/Output via MADI, Milan-AVB, or AES.
Unrivaled Performance
- Up to 40 dB of non-voice signal reduction.
- Consistently rated the highest quality output in blind tests.
- Hundreds of hours of continuous testing with no reduction in audio quality.
Optimized for Live Sound
- Latency of under 8.5 ms.
- Designed for seamless integration in L-Acoustics systems only.
- Extensively validated on major tours and productions, with no perceived impact on performance.
Source Intelligence in the News
常见问题解答
When answering this question, it’s important to remember that “zero latency” is a marketing term rather than a scientific fact. True “zero latency” does not exist in live audio because it is physically impossible to process audio signals instantly. End-to-end, the latency introduced by Source Intelligence is under 8.5ms over MADI. During its development, we ran extensive real-world tests with some of the most high-profile and demanding live sound engineers in the world. Their feedback was unanimous and clear – that this was within tolerable limits for use at FOH, given the exceptionally high quality of the resulting vocal.
Sound Intelligence greatly reduces the risk of feedback by removing non-voice signals by up to 40dB, including ringing, which can often turn into feedback. Source Intelligence achieves this by continuously processing the incoming signal and applying a proprietary machine learning algorithm to identify and isolate the vocal, removing all extraneous sound from the signal.
Sound Intelligence’s algorithm can also remove other non-vocal signals by up to 40dB, including bleed from the PA, room reverb, crowd noise, or even aggressive on-stage sounds such as live drums or guitar. There are no other products on the market with comparable performance.
Across multiple blind listening tests against leading noise suppression tools, Source Intelligence was consistently rated the best-sounding option, delivering significantly greater noise reduction while maintaining the same voice quality.
License activation is performed identically to L-ISA Controller, via the main menu in the software, select ‘Show License info’ and enter your Activation key. To remove a license from a computer, in this same menu, select ‘Release License.’
This product runs on a reliable hardware platform that has been thoroughly tested. We’ve run an extensive test grid that covers both audio quality and reliability. Source Intelligence has been lab-tested and extensively field-tested, including processing the main artist on multiple A-List arena and amphitheater tours, and tested at major events before launch.
Yes, there is PSU redundancy, Network Audio redundancy, and redundant control of multiple processors, offering full DSP redundancy.
To validate the system, the user must scan the network and identify at least 3 L-Acoustics Amplified Controllers. This can be performed from within the L-ISA Controller. Any current Amplified Controller is accepted: LA4X, LA12X, LA2Xi, LA7.16(i), LA1.16i
Yes, Source Intelligence requires a firmware update, but it will include L-ISA, so it’s the same as any other L-ISA update. This one update will include all updates for Source Intelligence and L-ISA.
No, at this time, an L-ISA Processor II can run either Source Intelligence or L-ISA, but not both simultaneously.
At this time, it is focused on voice; a customer can place this in the path of a single vocal channel, or on a group wherein multiple vocal channels feed it, such as in a corporate workflow: a group for handhelds, a group for lavaliers, and a group for headsets, allowing them to process more than 4 individual voices, but through 4 channels.