Back to Details

ace-studio-workflows

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

ACE Studio workflows

ACE Studio 工作流

Song generation

歌曲生成

When the user wants a whole song generated, use Inspire Me.
当用户希望生成整首歌曲时，使用 Inspire Me。

1. Generate

1. 生成

See 
ace-studio-features
for what Inspire Me takes and what it costs.
有关 Inspire Me 所需的输入及其费用，请参阅 
ace-studio-features
。

2. Align the project tempo to the result

2. 将项目速度与生成结果对齐

Do this before any editing — an unaligned project makes every later edit fight the grid. Which way you go depends on what else is on the timeline, because applying a tempo replaces the project's tempo and meter wholesale and moves existing content.
  • The generated audio is the only thing on the timeline. Nothing can be disturbed, so apply its tempo to the project directly.
  • The timeline already has other content. Ask the user first: confirm which version of the music they are keeping, and that it is the one the tempo should follow. Then apply that one's tempo. Everything already on the timeline will shift, so say so before you do it.
应在任何编辑之前完成此操作——未对齐的项目会让后续每次编辑都难以贴合网格。具体采用哪种方式取决于时间线上还有哪些内容，因为应用速度会整体替换项目的速度和拍号，并移动现有内容。
  • 时间线上只有生成的音频。 没有任何内容会受到影响，因此可直接将其速度应用到项目。
  • 时间线上已有其他内容。 先询问用户：确认要保留哪个版本的音乐，以及项目速度应跟随该版本。然后应用该版本的速度。时间线上已有的一切都会发生位移，因此执行前应告知用户。

3. Offer to keep going

3. 询问是否继续

Ask whether they want to edit the song. If they do, the first workflow below is the route in — split it into stems and transcribe the parts they want to change.
询问用户是否希望继续编辑这首歌曲。如果要编辑，下面第一个工作流就是切入方式——将歌曲拆分为 stems，并转录需要更改的部分。

Edit a song from audio

从音频编辑歌曲

When the user hands over a complete piece of music and wants to edit it, work through it like this.
当用户提供一首完整的音乐作品并希望编辑时，按以下流程处理。

1. Align the project to the audio

1. 将项目与音频对齐

Imported audio triggers beat analysis automatically. Apply that analysis so the project tempo matches the audio's tempo — the operation also snaps the audio's first downbeat forward to the nearest bar line, which shifts the audio later.
Tell the user the audio will move before you apply, and remember the whole apply replaces the project's tempo and meter wholesale.
导入音频后会自动触发节拍分析。应用分析结果，使项目速度与音频速度一致——该操作还会将音频的第一个强拍向前吸附到最近的小节线，从而使音频整体后移。
应用前告知用户音频会移动，并记住整个应用过程会整体替换项目的速度和拍号。

2. Split the music into stems

2. 将音乐拆分为 stems

Use the Stem Splitter. Pick the mode from what the user actually needs — see 
ace-studio-features
for what each mode separates and what it costs.
使用 Stem Splitter。根据用户的实际需求选择模式——每种模式所分离的内容及其费用请参阅 
ace-studio-features
。

3. Turn the melodic stems into MIDI

3. 将旋律 stems 转为 MIDI

3.1 The lead vocal

3.1 主唱人声（lead vocal）

Use the built-in Vocal to MIDI. It gives you the melody and the lyrics, which is why it beats a generic transcriber here. It needs a monophonic vocal.
使用内置的 Vocal to MIDI。它可同时提供旋律和歌词，因此在这里优于通用转录器。它需要单声部人声。

3.2 The other instrument stems — Basic Pitch, outside Studio

3.2 其他乐器 stems——Basic Pitch，在 Studio 之外

There is no instrument transcriber on ACE Studio's surface. Use Basic Pitch, which runs locally and free, then bring the MIDI back in.
The stems live on tracks, and Basic Pitch reads files — so the chain is: export each stem track as audio → run Basic Pitch on the file → import the resulting MIDI onto a track.
Setting it up. Check for a working environment first; do not install over the user's existing one. Basic Pitch wants an older interpreter than the newest system Python, so give it its own virtual environment — Python 3.11 is a compatible starting point:
python3.11 -m venv work/transcribe-env
work/transcribe-env/bin/python -m pip install basic-pitch librosa pretty_midi mido
On Windows the interpreter is 
work/transcribe-env/Scripts/python.exe
.
basic-pitch <output-dir> <stem.wav>
Basic Pitch is a candidate generator, not a transcription. It is built for single-instrument material, which is exactly what a stem is — but the output still needs cleaning before it is worth importing. Running it several times and getting the same answer is not evidence; it is the same model twice.
ACE Studio 界面中没有乐器转录器。使用 Basic Pitch，它可在本地运行且免费，然后将 MIDI 导回。
stems 位于轨道上，而 Basic Pitch 读取文件——因此流程为：
将每个 stem 轨道导出为音频 → 对该文件运行 Basic Pitch → 将生成的 MIDI 导入到轨道。
环境设置。 先检查是否已有可用环境；不要直接在用户现有环境中安装。Basic Pitch 所需的解释器比最新系统 Python 更旧，因此应为其创建独立的虚拟环境——可以从兼容的 Python 3.11 开始：
python3.11 -m venv work/transcribe-env
work/transcribe-env/bin/python -m pip install basic-pitch librosa pretty_midi mido
在 Windows 上，解释器为 
work/transcribe-env/Scripts/python.exe
。
basic-pitch <output-dir> <stem.wav>
Basic Pitch 是候选音符生成器，而不是真正的转录器。 它面向单乐器素材，而 stem 正是此类素材——但在导入前，输出仍需清理。多次运行并得到相同结果并不能作为证据；那只是同一个模型运行了两次。

3.3 Clean the MIDI before importing

3.3 导入前清理 MIDI

Remove what the recogniser invented:
  • Harmonic stacking — phantom notes an octave or a fifth above the real one.
  • Ultra-short noise notes and tail fragments left by decay.
  • Duplicate onsets on the same pitch at the same moment.
Keep what is musical: ornaments, syncopation, and rests are content, not noise. Judge against the actual tempo and phrasing — do not delete every short note with one fixed millisecond threshold, or you will flatten the performance.
Then check per instrument:
  • Bass. Octave errors are the common failure, and slides come back as dense runs of semitones. Cross-check the root against the kick and the chord changes. Keep it monophonic by default: bleed from guitar or keys must not become bass chords.
  • Guitar. Distorted guitar transcribes badly. Check short phrases against the spectrogram, the onsets, and the chord underneath.
  • Keys. Sustain pedal and tail resonance come back as masses of held notes — do not transcribe them literally. Reconstruct the two-hand division from the bass and the harmony.
  • Drums. Do not use Basic Pitch. It is a pitch model; percussion is out of scope for it.
删除识别器臆造的内容：
  • 和声堆叠——在真实音上方高八度或五度生成的幻觉音符。
  • 超短噪声音符以及衰减留下的尾音片段。
  • 同一音高在同一时刻出现的重复 onset。
保留具有音乐意义的内容：装饰音、切分节奏和休止符是作品内容，而非噪声。
应根据实际速度和乐句进行判断——不要用固定的毫秒阈值删除所有短音符，否则会抹平演奏细节。
然后按乐器进行检查：
  • 贝斯。 最常见的问题是八度错误，滑音通常会呈现为密集的半音连续音程。根据 kick（底鼓）和和弦变化交叉核对根音。默认应保持单声部：来自吉他或键盘的串音不得变成贝斯和弦。
  • 吉他。 失真吉他的转录效果很差。请结合频谱图、onsets 和下方的和弦检查短乐句。
  • 键盘。 延音踏板和尾音共振会变成大量持续音符——不要逐字照搬转录结果。应根据贝斯线和和声重建双手的声部分配。
  • 鼓组。不要使用 Basic Pitch。 它是音高模型，不适合处理打击乐器。

3.4 Load sound sources

3.4 加载音源

Load a suitable sound source onto whichever tracks the user wants to edit. See 
ace-studio-features
for choosing between AI Instruments and plugins.
在用户希望编辑的轨道上加载合适的音源。有关如何在 AI Instruments 与 plugins 之间选择，请参阅 
ace-studio-features
。

Export the project as sheet music

将项目导出为乐谱

When the user wants a printable score of what is in the project.
当用户需要项目内容的可打印乐谱时。

1. Get the notes out of Studio

1. 从 Studio 中获取音符

Export MIDI. It writes on the calling thread, so the file is on disk the moment the command returns — there is nothing to wait on. It carries every track that holds notes: voice, instrument, generic MIDI and chord tracks.
Audio tracks carry no notes, so they will simply be absent from the score. Check the project for them before you export — a bass or guitar that only exists as audio will vanish silently, and the user will not see it missing until they read the page.
To bring one in, run it through the transcription path from the first workflow: if it is a mixed audio track, split it into stems first; then export the audio, run Basic Pitch on the file, clean the result, and import it back as a MIDI track. For an audio track that is a lead vocal, use Vocal to MIDI instead — it brings the lyrics with it, which is what the score wants under the vocal line. Once the material is on note-carrying tracks, export MIDI again.
Ask the user before doing this: transcription is an estimate, and they may prefer a score of only the parts they actually authored.
Know what MIDI drops. Lyrics, syllables, vocal parameters and articulations have nowhere to live in a MIDI file. If the score needs lyrics under the vocal line, export the vocal separately as UfData — which carries them, but only for voice tracks — or plan to type them into the notation program.
There is no MusicXML export on this surface, so MIDI is the way across.
导出 MIDI。它会在调用线程上执行写入，因此命令返回时文件已在磁盘上——无需等待。它会包含所有包含音符的轨道：人声、乐器、通用 MIDI 和和弦轨道。
音频轨道不包含音符，因此不会出现在乐谱中。 导出前检查项目中是否存在音频轨道——仅以音频形式存在的贝斯或吉他会悄然消失，而用户直到阅读乐谱时才会发现缺失。
若要将音频轨道导入，请按照第一个工作流中的转录流程处理：如果是混合音轨，先将其拆分为 stems；然后导出音频，对文件运行 Basic Pitch，清理结果，再作为 MIDI 轨道导回。对于属于 lead vocal 的音轨，改用 Vocal to MIDI——它还会带来歌词，这正是谱面在人声声部下所需的内容。素材置于含音符的轨道上后，再次导出 MIDI。
执行前询问用户：转录结果是估算值，用户可能更希望只为实际创作的部分制谱。
了解 MIDI 会丢失的内容。 歌词、音节、人声参数和发音法在 MIDI 文件中无处存储。如果乐谱需要在人声声部下显示歌词，请将人声单独导出为 UfData——它可以携带歌词，但仅适用于人声轨道——或者计划在制谱软件中手动输入。
此界面不支持 MusicXML 导出，因此应通过 MIDI 过渡。

2. Notate it in MuseScore

2. 在 MuseScore 中制谱

MuseScore Studio is the free notation and export tool for this. Find the executable rather than assuming a path: check 
PATH
, the executable path of a running process, the usual OS install directories, or ask the user. On Windows the file is usually 
MuseScore4.exe
— do not assume it is on the C: drive.
Import the MIDI, build the score, and save a native 
.mscz
. Export PDF and MIDI from that 
.mscz
, so all three files come from one source:
<musescore> -o work/score.pdf work/score.mscz
<musescore> -o work/score.mid work/score.mscz
-o
picks the format from the target extension. Do not pass 
-F
 — it clears the user's settings — and do not reach for 
--force
to paper over a score that will not open.
MuseScore Studio 是免费的制谱与导出工具。请查找可执行文件，不要直接假定路径：检查 
PATH
、正在运行进程的可执行文件路径、操作系统常用安装目录，或询问用户。在 Windows 上，该文件通常是 
MuseScore4.exe
——不要假定它位于 C: 驱动器。
导入 MIDI、构建乐谱，并保存为原生 
.mscz
文件。从该 
.mscz
文件导出 PDF 和 MIDI，使这三个文件都来自同一个源文件：
<musescore> -o work/score.pdf work/score.mscz
<musescore> -o work/score.mid work/score.mscz
-o
根据目标扩展名选择格式。不要传入 
-F
——它会清除用户设置——也不要使用 
--force
来掩盖无法打开的乐谱。

3. Clean up what MIDI import produces

3. 清理 MIDI 导入产生的内容

A MIDI import is a starting point, not a score. Expect to fix:
  • Note durations and rests. A performance MIDI has notes that end slightly early or late; notation needs real durations, ties and rests.
  • Enharmonics and key. MIDI carries pitch numbers, not spellings, so the key signature and accidentals need setting by hand.
  • Voice and staff assignment. Keyboard parts arrive as one stream and need splitting across two staves; a percussion track may import as piano.
  • Tempo map. Confirm the imported tempo and meter match the project, especially if beat analysis was applied earlier.
MIDI 导入只是起点，并不是成品乐谱。需要修正以下内容：
  • 音符时值与休止符。 演奏型 MIDI 中的音符结束时间可能略早或略晚，而制谱需要准确的时值、连音线和休止符。
  • 等音与调号。 MIDI 只携带音高数字，不携带音名拼写，因此需要手动设置调号和临时记号。
  • 声部与谱表分配。 键盘声部导入后是单一音符流，需要拆分到两个谱表；打击乐轨道可能会作为钢琴导入。
  • 速度映射。 确认导入的速度和拍号与项目一致，尤其是在此前应用过节拍分析的情况下。

4. Verify before you hand it over

4. 交付前验证

Do not report a score as finished on the basis that the files exist.
  1. Reopen the final 
    .mscz
    in MuseScore and confirm it is editable.
  2. Parse the exported MIDI (
    mido
    or 
    pretty_midi
    ): check the tempo map, that notes are present, the voice mapping, and the durations. Look for octave shifts, drums exported as piano, and empty tracks. A non-empty file with a valid header proves only that it is a MIDI file.
  3. Render every page of the PDF to an image (
    pdftoppm -png -r 120 …
    ) and check each one for clefs, beams, ties, rhythm and layout. Spot-checking a couple of pages is not the final check.
  4. If you edit the 
    .mscz
    after exporting, re-export both PDF and MIDI from it.
Fix problems in the source score and re-check whatever the fix touched. Handle routine issues yourself; do not ask the user about every spacing adjustment.
Describe the result honestly: say what you verified and what you did not. Call it a playable arrangement unless you actually have evidence for note-level fidelity.
不要仅凭文件存在就报告乐谱已完成。
  1. 在 MuseScore 中重新打开最终的 
    .mscz
    文件，确认其可编辑。
  2. 解析导出的 MIDI（
    mido
    或 
    pretty_midi
    ）：检查速度映射、音符是否存在、声部映射和时值。查找八度偏移、被导出为钢琴的鼓组以及空白轨道。文件非空且文件头有效，只能证明它是一个 MIDI 文件。
  3. 将 PDF 的每一页渲染为图像（
    pdftoppm -png -r 120 …
    ），并逐页检查谱号符、符梁、连音线、节奏和版式。抽查几页不能作为最终检查。
  4. 如果在导出后编辑了 
    .mscz
    文件，请重新从中导出 PDF 和 MIDI。
在源乐谱中修复问题，并重新检查所有受修复影响的内容。自行处理常规问题；不要就每个间距调整询问用户。
如实描述结果：说明验证了什么，以及未验证什么。除非确实有音符级保真度方面的证据，否则应将其称为可演奏的编配。