ace-studio-features

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

ACE Studio features

ACE Studio 功能

Tracks

轨道

Track types

轨道类型

A track has no fixed type, and by default it is a blank track. It becomes what you put on it — load a singing voice and it is a vocal track, load an instrument and it is an instrument track. A project starts with a large pool of empty tracks waiting to be filled.

	Catalog	Driven by
Blank track	Blank
Audio track	Audio	audio clip
Voice track	MIDI	MIDI + lyrics + vocal parameters
AI instrument track	MIDI	MIDI + articulations + instrument parameters
Plugin instrument track	MIDI	MIDI + MIDI CC
Master track

A MIDI track with no sound source is not silent — it falls back to a standard piano sample library. When the user says "it sounds like a piano", the answer is almost always that nothing is loaded, not that something broke.
Every track except a blank track can host audio effects.
Unison Mode belongs to voice tracks and AI instrument tracks only. It lets several voices or instruments load onto one track to produce a unison. Open the spread up when you want a wider stereo image.

轨道没有固定类型，默认是空白轨道。它会随你放入的内容而改变——加载歌唱声音后，它就是人声轨道；加载乐器后，它就是乐器轨道。项目启动时会有大量等待填充的空白轨道。

	目录	驱动方式
空白轨道	空白
音频轨道	音频	音频片段
人声轨道	MIDI	MIDI + 歌词 + 人声参数
AI 乐器轨道	MIDI	MIDI + 演奏法 + 乐器参数
插件乐器轨道	MIDI	MIDI + MIDI CC
主轨道

一个没有声音源的 MIDI 轨道并非处于静音状态——它会回退到标准钢琴采样库。当用户说“它听起来像钢琴”时，几乎总是因为没有加载任何声音，而不是哪里出了故障。
除空白轨道外，每条轨道都可以承载音频效果。
Unison Mode 仅适用于人声轨道和 AI 乐器轨道。它允许多个人声或乐器加载到同一条轨道上，形成齐奏。想获得更宽的立体声像时，可以增大 spread。

Audio effects

音频效果

ACE Studio ships basic effects, but prefer the third-party plugin effects the user has installed locally.
Voice tracks and instrument tracks come with a basic effect chain already loaded. Add to it or strip it back as the work requires.
Room Effect is a spatial effect that exists to serve Unison Mode: it models one acoustic space and places the unison speakers at different positions within it. Turn it on only when the track plays a background role.
To take an effect out of the chain, deactivate (disable) it — do not bypass it. Bypass is for comparing a plugin in context, a listening call an agent does not make.

ACE Studio 内置基础效果，但应优先使用用户本地安装的第三方插件效果。
人声轨道和乐器轨道默认已经加载一条基础效果链。可根据作品需要添加效果，或将其精简。
Room Effect 是一种为 Unison Mode 服务的空间效果：它会建模一个声学空间，并将齐奏声源置于该空间中的不同位置。仅当该轨道承担背景角色时才开启。
若要将某项效果移出效果链，请停用（禁用）它，而不是使用旁路（bypass）。 Bypass 用于在完整上下文中比较插件的效果，这是 Agent 不会作出的听感判断。

Vocal Synth, AI Instruments, and instrument plugins

Vocal Synth、AI Instruments 和乐器插件

Vocal Synth

Use Vocal Synth when the user wants to create a vocal. It is the sound source that takes MIDI and lyrics, and synthesises a sung vocal through a chosen speaker.

Recommended speakers:

For	Speakers
Big pop vocals	Ember Rose, Rebecca, Bianca, Mangus
R&B, jazz	Jessica, Zalo
Lyrical, tender	Elirah, Emma
Epic, cinematic	Clara

Blend voice creates a new voice by mixing several speakers. When the user describes a voice they want, pick from the library first; reach for a blend only if the library does not get there and they want something bespoke.

If the user wants their own voice, or a specific voice they supply, open the voice library and point them at the Cloning tab to train it themselves. That flow happens in a browser. Consult the wiki and walk them through it.

Vocal Synth renders in the cloud, but there is a turbo mode that moves most of the model computation onto the local machine. Studio measures whether the machine can support turbo mode the first time it launches: if the local compute is enough it is on by default, and if the check fails it cannot be enabled. Turbo mode turns on pre-rendering by default, so an edit triggers a render immediately. With turbo mode and pre-rendering together, rendering is usually finished before playback even starts.

Without pre-rendering, rendering is triggered by playback.
With pre-rendering, rendering is triggered by the edit.
During playback, a region that has not finished rendering blocks playback.

当用户想创建人声时，使用 Vocal Synth。它是接收 MIDI 和歌词、并通过选定的歌手合成歌唱人声的声音源。

推荐歌手：

适用场景	歌手
宏大的流行人声	Ember Rose、Rebecca、Bianca、Mangus
R&B、爵士	Jessica、Zalo
抒情、柔和	Elirah、Emma
史诗、电影感	Clara

Blend voice 通过混合多个歌手来创建新声音。当用户描述想要的声音时，先从声音库中挑选；只有当现有声音库无法满足需求、且用户希望获得定制声音时，才使用 blend。

如果用户想使用自己的声音，或使用他们提供的特定声音，请打开声音库，并指引他们前往 Cloning 标签页自行训练。该流程在浏览器中完成。请查阅 wiki，并一步步引导他们完成。

Vocal Synth 在云端渲染，但提供 turbo mode，可将模型的大部分计算转移到本地计算机。Studio 首次启动时会检测该计算机是否支持 turbo mode：如果本地算力足够，默认开启；如果检测失败，则无法启用。Turbo mode 默认开启 pre-rendering，因此每次编辑都会立即触发渲染。Turbo mode 与 pre-rendering 同时启用时，渲染通常会在播放开始前就已完成。

未启用 pre-rendering 时，渲染由播放触发。
启用 pre-rendering 时，渲染由编辑触发。
播放期间，尚未完成渲染的区域会阻塞播放。

AI Instrument

When the user wants an acoustic instrument, prefer an AI Instrument — its realism is far beyond a sample library.

AI Instruments do not render in real time, so each instrument type carries a very small sample pack purely so MIDI input responds as you play. That sample is not what the model outputs.

String Section is modelled on a string orchestra. Its speakers are Violins I, Violins II, Violas, Celli and Basses. These support polyphony well, but keep the written part idiomatic for the section it is on.

AI Instruments render in the cloud, with no turbo mode and no pre-rendering.

Rendering is triggered by playback.
During playback, a region that has not finished rendering blocks playback.

Every other instrument is a solo instrument. You can feed them polyphonic material, but they do not behave the way a sample library does:

Strings tolerate a little polyphony — but as one instrument bowing several strings, not as a section.
Wind instruments cannot physically play polyphonically, so the model is forced to fake it. Do not do this.
When a solo instrument needs several parts, split them across tracks instead.

Instruments named Vintage perform less well than the others.

Articulations are marked on notes, and each instrument type has its own set. The default smart mode works out the articulation from the melody and applies it in the synthesised result. Two articulations never appear under smart mode and must be set explicitly:

Pizzicato
Mute

当用户需要原声乐器时，优先使用 AI Instrument——其真实感远超采样库。

AI Instruments 不进行实时渲染，因此每种乐器类型都携带一个非常小的采样包，纯粹用于在演奏时让 MIDI 输入能够即时响应。该采样并不是模型的输出。

String Section 以弦乐团为建模基础。其声部包括 Violins I、Violins II、Violas、Celli 和 Basses。这些声部能够很好地支持复调，但书写的声部部分应采用适合该声部的惯用写法。

AI Instruments 在云端渲染，不支持 turbo mode，也不支持 pre-rendering。

渲染由播放触发。
播放期间，尚未完成渲染的区域会阻塞播放。

其余每一种乐器都是独奏乐器。 你可以为其输入复调素材，但它们的行为方式与采样库不同：

Strings 可以承受少量复调——但应当像一名演奏者用弓演奏多根弦，而不是一个弦乐声部。
管乐乐器在物理上无法进行复调演奏，因此模型只能被迫模拟。不要这样做。
当一个独奏乐器需要演奏多个声部时，请将它们拆分到不同轨道上。

名称中带有 Vintage 的乐器表现不如其他乐器。

演奏法标记在音符上，每种乐器类型都有自己的一组演奏法。默认的 smart mode 会根据旋律判断演奏法，并将其应用到合成结果中。smart mode 下永远不会出现以下两种演奏法，必须显式设置：

Pizzicato
Mute

Instrument plugins

乐器插件

ACE Studio supports third-party VST3 and AU instrument plugins.

When the user wants a particular sound:

If it is an acoustic instrument, look through the AI Instrument list first.
Then look through the user's own plugin list for something suitable.

Once a plugin is loaded, it is fully drivable — parameters, whole state and presets, and its own interface through Studio's built-in computer use. See

ace-studio-audio-plugins

for how to drive it, and for per-plugin recipes.

ACE Studio 支持第三方 VST3 和 AU 乐器插件。

当用户需要特定声音时：

如果是原声乐器，先查看 AI Instrument 列表。
然后在用户自己的插件列表中查找合适的插件。

插件加载后即可完全驱动——包括参数、完整状态和预设，以及通过 Studio 内置的 computer use 操作插件自身的界面。 有关如何驱动插件以及各插件的具体用法，请参阅

ace-studio-audio-plugins

。

Generative Kits

生成式工具包

Every generative kit takes a prompt and produces audio. Generation takes over a minute. There are three:

Inspire Me — text to song. Use this when the user wants to generate a whole song outright.
Music Enhancer — AI cover. Select a region of the timeline as the input context. Once the tool has uploaded the audio it returns the lyrics and style it recognised; edit either (or neither) to produce the cover. Three common uses:
- Style replacement. Keep the lyrics, rewrite the style description, turn one finished piece of music into another.
- Finishing an idea. Record a hum or write a melodic motif, then write the lyrics and style out in full to grow the motif into a complete demo.
- Refinement. Keep the lyrics and style as they are and regenerate. The content usually shifts slightly, but it turns a rough demo into a polished one.
Add A Layer — generation as splice. Select the region to generate into. The tool uploads the surrounding timeline content as reference audio and generates the specified sample from the prompt. The result fits what is already on the timeline — a bass line, a drum part, a lush string passage.

每个生成式工具包都接收提示词并生成音频。生成过程耗时 超过一分钟。共有三种：

Inspire Me — 文本生成歌曲。当用户想直接生成整首歌曲时使用。
Music Enhancer — AI 翻唱。选择时间线上的一个区域作为输入上下文。工具上传音频后，会返回识别出的歌词和曲风；可编辑其中之一（或两者都不编辑），以生成翻唱。常见用途有三种：
- 曲风替换。 保留歌词，重写曲风描述，将一首完成的作品转换成另一种风格。
- 完善创意。 哼唱一段或写下旋律动机，再完整写出歌词和曲风，将该动机扩展成完整的 demo。
- 优化。 保持歌词和曲风不变并重新生成。内容通常会略有变化，但它能把粗糙的 demo 打磨成精致的成品。
Add A Layer — 拼接式生成。选择要生成到其中的区域。工具会将周围时间线内容作为参考音频上传，并根据提示词生成指定采样。结果会与时间线上已有内容相匹配，例如贝斯线、鼓声部或华丽的弦乐段落。

How to prompt Add A Layer

如何为 Add A Layer 编写提示词

Choose custom mode. Then write two short prompts, one per field:

Instrument — the instrument and its timbre.
Styles — the musical content: style, emotional colour, layering and dynamics, and the groove and physical feel of the playing.

Keep both short. Do not describe rhythm or chords — the surrounding audio is uploaded as reference, so the model already has the harmony and the groove; restating them adds nothing. And do not pile up negative prompts: this model is built for synthesising a single timbre, so by default it does not need to be told what to avoid.

选择 custom mode（自定义模式）。 然后分别填写两个简短提示词，每个字段一个：

Instrument — 乐器及其音色。
Styles — 音乐内容：风格、情感色彩、层次与动态，以及演奏的 groove 和实际手感。

两者都应保持简短。不要描述节奏或和弦——周围音频已作为参考上传，因此模型已经掌握和声与 groove；重复描述不会增加任何信息。也不要堆砌负面提示词：该模型专为合成单一音色而构建，默认情况下无需告诉它应避免什么。

Utilities

实用工具

Stem Splitter

Four ways to split:

Vocal, instrumental.
Vocal, drums, bass, piano, guitar, others. Nothing is distinguished within a category — several guitar tracks may all land in the guitar stem — and some instruments end up in Others.
All detected stems. Separation is not guaranteed to be complete; instruments of the same category often share a stem, and some land in FX or Others.
Customized. Describe what to separate in a prompt. This suits non-musical audio better; on musical content the granularity is coarse — ask for hi-hat and you will likely get the whole drum kit. The operation produces two tracks: the separated target, and everything left over.

有四种拆分方式：

人声、器乐伴奏。
人声、鼓、贝斯、钢琴、吉他、其他。 同一类别内不做进一步区分——多条吉他轨道可能全部进入 guitar stem，有些乐器则最终进入 Others。
所有检测到的 stems（分轨）。 无法保证分离完整；同类别乐器通常会共用一个 stem，有些则进入 FX 或 Others。
自定义。 在提示词中描述要分离的内容。这种方式更适合非音乐音频；对于音乐内容，分离粒度较粗——要求分离 hi-hat 时，很可能得到整套鼓组。该操作会生成两条轨道：分离出的目标，以及其余所有内容。

Voice Changer

Converts a monophonic vocal on the timeline into a vocal or instrument audio in a chosen timbre.

A set of preset models is included to cover a wide range of creative needs, and the surface lists every model a conversion can use. When the user needs a timbre outside it, guide them to train their own Voice Changer model.

Training input is usually a vocal, but in practice it also accepts instrument audio with a clear timbre — which must also be monophonic.

将时间线上的单声部人声转换为具有指定音色的虚拟人声或乐器音频。

内置一组预设 models，覆盖广泛的创意需求，界面会列出转换时可用的所有模型。当用户需要列表之外的音色时，引导他们训练自己的 Voice Changer 模型。

训练输入通常是人声，但实际也可接受音色清晰的乐器音频——该音频同样必须是单声部的。

Vocal to MIDI

Recognises a sung vocal in audio as a MIDI clip with lyrics. Supports monophonic vocals; it cannot correctly resolve stacked, multi-part vocals.

The V2 model can pull the lead vocal's lyrics and melody straight out of a complete song, and holds up well against background noise.

Expect some notes to come back at the wrong pitch, and fix them before the clip is used. A single note is not enough to judge — a drifting note and a deliberate accidental look the same in isolation. Read the melody as a whole, work out what key it is actually in, and correct the notes that fall outside it.

将音频中的歌唱人声识别为带歌词的 MIDI 片段。支持单声部人声；无法正确解析叠录的多声部人声。

V2 模型可以直接从完整歌曲中提取主唱人声的歌词和旋律，并且对背景噪声有较好的抵抗力。

预计会有部分音符以错误音高返回，并应在使用片段前修正。 单个音符不足以判断——偏移的音符和刻意的变化音单独听起来完全一样。应整体阅读旋律，判断实际调性，并修正落在调性之外的音符。

Doubles

Vocal Synth tracks and AI Instrument tracks support doubles.

The result is two duplicated tracks whose notes carry a degree of random offset from the originals. The two new tracks are panned left and right.

Vocal Synth 轨道和 AI Instrument 轨道支持 Doubles。

结果会生成两条复制的轨道，其音符相对原始音符带有一定程度的随机偏移。两条新轨道分别声像置左和置右。

Marker track

Marker track（标记轨）

A marker track can be created at the top of the timeline to give the user reference annotations against time — sections, lyrics, and so on.

可以在时间线顶部创建 Marker track，为用户提供与时间对应的参考注释——例如段落、歌词等。

Tempo track

Tempo track（速度轨）

When the music changes tempo, draw tempo automation on the tempo track. Note that editing the tempo track overwrites globally.

When the user wants the project tempo synced to imported audio, use the Apply Tempo interface to apply the audio's tempo to the project. Note that this operation shifts the target audio later so that the audio's first downbeat lands on the nearest bar line.

当音乐速度发生变化时，在 Tempo track 上绘制速度自动化曲线。请注意，编辑 Tempo track 会进行全局覆盖。

当用户希望将项目速度与导入的音频同步时，使用 Apply Tempo 界面，将该音频的速度应用到项目中。请注意，此操作会将目标音频向后移动，使音频的第一个强拍落在最近的小节线上。

Importing files

导入文件

Studio imports the common formats onto the timeline — MIDI, audio, MusicXML and video among them. The surface's docs carry the full extension list.

Studio 可将常见格式导入时间线，包括 MIDI、音频、MusicXML 和视频等。操作界面的文档中提供了完整的扩展名列表。