审阅范围 · Review scope
限定范围独立审阅。请保留正文中的出处与限制；不代表全库最终验收、实际演奏或学习效果验证。
Independent review within the stated scope. Keep the evidence and limitations stated in the text in view. This is not final corpus acceptance or validation of performance or learning outcomes.
先分清低音与音组，再练交替伴奏
本包通过限定独立文字／静态来源、双语含义、初学者说明与符号审阅，未发现实质问题。T2是指定手册／静帧支持的作者归属，G是本例限定术语，O是本库原创纸面任务与练习规划。来源逐项布尔值保留作者冻结时快照，publication_review记录当前通过范围；不表示来源听校、连续动作、完整来源谱例、演奏或学习效果验收。
看到伴奏的一团音，先别把它当成“每拍都要抓一个大和弦”。先问低音何时起、上面的音组何时起、哪些拍留空，再安排动作。一个节拍位置同时弹两个音，仍然只占那个时间位置；音符数量、起音时点与小节长度是三件事。
已读 IV-021 全部62段既有ASR，检查12次选定静帧及讲义实际／印刷31–33页完整文字与选定标签、起音布局。31段候选因未核历史／音响／动作概括、重复尾段或时间异常等不作归属证据，不是31个已听校错误。完整原谱、原声与连续动作未验收。
纸面路线可独立完成，不需要听音或钢琴。可选弹奏仅在能找键、数拍并舒服地逐次按放单音／两音时进行；RH 是右手，LH 是左手。本页把低音交给LH、上方两音交给RH，不是源课单手跳跃练习的动作替代验证。无需踏板、固定指法或大跨度；不舒服就保留纸面路线。
G：低音是较低的声部角色；根音是给和弦命名的参照，二者不必相同。这里“音组”是同一起点按下的一组音；“声部”是持续追踪的一条音线，不等于一只手。起音是开始一个音，保持是继续按住，休止是该声部没有保持音；没再起音不必然等于休止。
有边界的来源归属
B45-S01 · T2
第21课把低音与上方音组交替安排，并展示低音放在第一拍、或第一与第三拍，以及换用其他和弦音作低音的变化。我们只借用“先列角色和时点”的思路，不复制源谱，也不把一种分配当作所有伴奏的规则。
PDF-IV实际／印刷31页标题与三组布局；IV-021静帧19、72、138、153秒的角色／起音提示。完整和声、全部音符及原演奏未逐项验收。
B45-S02 · T2
作者让学习者试不同音区和音的数量，再参考示范。部分“减少／增加音”的谱例同时换了和弦标签，不能概括成和声完全不变、只改厚薄的实验。音区与宽跨度建议保留为作者经验，本页不据此要求某种手型、跨度或踏板用量。
PDF-IV实际／印刷32页选定音组与标签、33页完整任务文字；IV-021静帧170／190秒音区建议、215／235秒音数及和弦标签、255秒跨度示例、275秒任务提示。没有听音比较，也未检查33页完整旋律。
把练习拆成角色、拍位、释放三个层次
G／O：下面原创材料以 F 大三和弦 F–A–C 为和声计划。低音用 F3，上方音组为 A3＋C4，两音同时起放；这个上方二音组不是完整三和弦，也不能单独证明 F 是主音。三种音在一小节内先后出现，可以表达我们的计划，不等于三音一直同时响。
G／O：3/4 每小节三拍、4/4 每小节四拍，本页都以四分音符为一拍。原创T版为“低音—音组—音组”；F版为“低音—音组—低音—音组”。它们是不同拍号下的书面比较，不是把源课或任意真实曲子直接改拍号，也不把三拍伴奏自动判成华尔兹风格。
G／O：V版只把F版第二次低音从 F3 改为 C3。C仍是F和弦的五度音，计划的根音仍为F；它在这个音区位于F3下方5个半音。“五度和弦音”是相对根音的类别，不表示这里向上跳了五度。实际分析一首曲子还需旋律与上下文，不能仅凭换了低音就断言换和弦或转调。
O：拿到真实乐谱时，先标原拍号、每个起点和声部释放，再选一个小节练。若谱上有延音、连线、踏板或不同声部，按原谱另建完整时间线；不能把本页半拍短音套上去。低音与和弦若原由同一手承担，本页两手分工只练先后顺序，未验证原来的跳跃动作。
原创纸面任务：同拍两音，算一次时点
先遮住答案。按下方角色表与每版时点表列出每条声部的保持／休止。T是3/4；F、V、R是4/4；这些字母是版本名，不是此处新增的和弦代号。所有版本四分音符=60 BPM，每个音保持半拍；每版活动小节之后再留一个同拍号的完整休止小节。R完全保留F的右手音组与8拍总长，只省去两次低音。
G／O：中央C为C4；F3是其左侧低一个八度内的F，A3是中央C左侧的A，C3比中央C低一八度。88键由A0=1算起：C3=28、F3=33、A3=37、C4=40。A3–C4相隔3个半音，键号不是指法。若不能舒服地同时按这两个键，可直接做纸面；改成先后两音会成为另一种任务，不计为这里的同时音组。
每版有几个音符起音事件、几个不同起音时点、总共几秒？在V里，变的是根音、低音、拍号还是上方音组？R恢复成F时，除了放回低音，还要改哪条休止记录？所有时点以四分音符为拍单位，从版本开始累计，0对应第一拍开头，不是“第零拍”。
| 角色 | 音高／键号 | 手／声部 |
|---|---|---|
| 低音 | F3 / 33 | LH / 低音 |
| 内声 | A3 / 37 | RH / 内声 |
| 上声 | C4 / 40 | RH / 上声 |
版本起点
| 版本 | 拍号 | 低音起点 | 上方双音起点 | 总拍数 | 恢复区间 |
|---|---|---|---|---|---|
| T | 3/4 | 0: F3 | 1, 2 | 6 | 3–6 |
| F | 4/4 | 0: F3, 2: F3 | 1, 3 | 8 | 4–8 |
| V | 4/4 | 0: F3, 2: C3 | 1, 3 | 8 | 4–8 |
| R | 4/4 | 无 | 1, 3 | 8 | 4–8 |
核对答案
T有5个音符事件、3个起音时点，总6秒；F／V各6个事件、4个时点、8秒；R有4个事件、2个时点、8秒。音组两音同时出现，所以不能按音符个数算总时长。V只改第二次低音的音高／键号，不改计划根音F、拍号或上方音组。R→F须恢复LH／低音声部0–0.5与2–2.5的保持，并从其整段休止中移除这些区间。
O：若双音被算成两拍，先把同起点事件圈在一起；若R加入低音后还标全休止，先修时间线再试弹。若你不知道原谱的声部释放，就标“待核”，不要自行用踏板填满。纸面核对可独立完成，不需要对风格或音色给出听觉结论。
可选练习：交替起放，保留空拍
准备好时先做完整R，只弹右手两次音组，左手休息；数完整两小节，不跳过前面的空拍。再按表恢复F的低音。T与V是另外的已写明比较，可选择做，不把全部连起来当一首原曲。保持舒适音量意图，不用踏板、不要求左右手同时跨握；每个音到半拍就释放，不把这里的时间规定当通用断奏定义。
每行是一个声部：bass=低音／LH，inner=内声／RH的A3，upper=上声／RH的C4。B1／B2是第一／第二次低音；Q1／Q2是第一／第二次双音，i／u分别指内／上声。每个方括号是一小节；格子各半拍，ID表示该音起并保持半拍，.表示该声部休止。两条右手行同格有ID，表示两音同时起放。所有版本最后一小节三条声部都休止。
T bass: [B1 . . . . .] [. . . . . .]
T inner: [. . Q1i . Q2i .] [. . . . . .]
T upper: [. . Q1u . Q2u .] [. . . . . .]
F bass: [B1 . . . B2 . . .] [. . . . . . . .]
F inner: [. . Q1i . . . Q2i .] [. . . . . . . .]
F upper: [. . Q1u . . . Q2u .] [. . . . . . . .]
V bass: [B1 . . . B2 . . .] [. . . . . . . .]
V inner: [. . Q1i . . . Q2i .] [. . . . . . . .]
V upper: [. . Q1u . . . Q2u .] [. . . . . . . .]
R bass: [. . . . . . . .] [. . . . . . . .]
R inner: [. . Q1i . . . Q2i .] [. . . . . . . .]
R upper: [. . Q1u . . . Q2u .] [. . . . . . . .]T的低音保持0.5拍，每条右手声部保持1拍；F／V低音保持1拍，每条右手声部仍1拍；R低音保持0拍。其余时间各声部均休止。T恢复小节为3–6，F／V／R为4–8；T最后音组在2.5释放，F／V／R在3.5释放。没有踏板或重叠保持，实际声学余响未测量。
R接回F，右手事件原样保留，加入B1与B2；LH／bass休止由0–8改成0.5–2和2.5–8。V接回F，只把B2的C3／28恢复为F3／33。T不是F的删音版：它另有3/4拍号、音组在1和2以及3–6恢复小节，不能只删除F的一拍而保留旧小节线。
休息后遮住表格，重建F的三个声部，并说明V为什么不自动改变根音，作为保留重测。回到真实乐句时，先恢复原谱的拍号、低音、时值和手部分工，再练接缝；是否能独立单手完成另行核对。未知乐谱迁移题需另备且不预先给答案，本包没有提供或验证。
边界与相关阅读
来源账本为380.054秒，同哈希容器为380.053秒，差异原因未定，不是实际听看时长；ASR最后一段终点还越过容器时长，已排除，未据此推断原因。没有原音频听校、连续动作、完整源谱、踏板效果、身体适用性、学习成效或应用验证。
本包来源范围
角色、事件和时间检查
Separate the bass and upper group before practicing alternation
This package passed bounded independent written/static-source, bilingual-meaning, beginner-instruction and symbolic review with no material findings. T2 marks attribution supported by specified handbook/still evidence, G scoped terms, and O original paper tasks and practice planning. Per-source booleans retain their author-time snapshot; publication_review records the current passed scope, not source audition, continuous motion, complete source scores, performance or learning-efficacy acceptance.
When accompaniment looks like a cluster, do not immediately assume a large chord must be grabbed on every beat. First ask when the bass attacks, when the upper group attacks and where time stays empty. Two notes attacked together still occupy one time position. Note count, attack times and bar length are different quantities.
All 62 existing ASR segments of IV-021 were read, with 12 selected stills and complete prose plus selected labels/attack layout on handbook physical/printed pages 31–33. 31 candidate segments are excluded from attribution for unverified historical/sounding/movement generalizations, repeated tail text or timing anomalies; these are not 31 audition-confirmed errors. Complete source notation, original audio and continuous movement remain unvalidated.
The paper route is independently complete without listening or a piano. Optional playing requires finding keys, counting beats and comfortably pressing/releasing successive single notes or pairs; RH means right hand and LH left hand. We assign bass to LH and the upper pair to RH, which does not validate the source lesson’s one-hand leaps. No pedal, fingering or wide span is required; keep the paper route if uncomfortable.
G: bass is the lower voice’s role; the root names the chord, and they need not coincide. A group here contains notes attacked at the same onset. A voice is a tracked musical line, not a hand. Attack starts a note, holding continues it, and a rest means that voice holds no note; absence of a new attack need not mean rest.
Bounded source attribution
B45-S01 · T2
Lesson 21 alternates bass with upper groups and shows bass on beat one or beats one/three, including a different chord tone as bass. We retain only the planning idea of listing roles and times, without reproducing its score or treating one distribution as a universal accompaniment rule.
PDF-IV physical/printed page 31 headings and three layouts; IV-021 stills at 19, 72, 138 and 153 seconds show role/attack prompts. Complete harmony, every note and the original performance were not individually accepted.
B45-S02 · T2
The author asks learners to try registers and note counts before consulting the demonstration. Some examples that remove/add notes also change chord labels, so they are not a harmony-preserving experiment in thickness alone. Register and wide-span advice remain author heuristics; this page derives no hand shape, span or pedal amount from them.
Selected groups/labels on PDF-IV physical/printed page 32 and complete task prose on page 33; IV-021 stills 170/190 show register advice, 215/235 show counts/chord labels, 255 shows span examples and 275 the task prompt. No sounding comparison or complete page-33 melody review was performed.
Separate roles, beat positions and releases
G/O: our original material declares the F-major triad F–A–C as its harmonic plan. Bass uses F3, and the upper group is A3+C4, attacked/released together. This upper pair is not a complete triad and cannot alone prove F as tonic. The three pitch classes occur successively within a bar; they do not sound together continuously.
G/O: 3/4 has three beats per bar and 4/4 four, with a quarter note as the beat here. Original T uses bass–group–group; F uses bass–group–bass–group. This is a written comparison under different meters, not permission to re-meter the source lesson or an arbitrary piece, nor proof that any three-beat accompaniment has waltz style.
G/O: V changes only F’s second bass from F3 to C3. C remains the fifth chord tone of F, whose planned root stays F; in this register C3 lies 5 semitones below F3. “Fifth chord tone” is a category relative to the root, not a claim that this event leaps upward by a fifth. Real analysis still needs melody and context; a changed bass alone does not prove a new chord or modulation.
O: with a real score, mark its meter, onsets and voice releases before selecting a bar. If it has holds, ties, pedal or different voices, construct its own complete timeline; do not impose our half-beat notes on it. If one hand originally plays both bass and chord, this two-hand division rehearses order only and does not validate the original leaps.
Original paper task: two notes together, one attack time
Hide the answers. Use the role and onset tables to list holds/rests for each voice. T is 3/4; F, V and R are 4/4. These letters name versions, not additional chord symbols here. Every version is at quarter note=60 BPM and holds each note for half a beat, followed by one whole rest bar in the same meter. R retains all of F’s RH groups and its 8-beat total, omitting only the two bass attacks.
G/O: middle C is C4; F3 is the F within the octave to its left, A3 the A below middle C, and C3 one octave below it. Counting 88 keys from A0=1 gives C3=28, F3=33, A3=37 and C4=40. A3–C4 spans 3 semitones; key numbers are not fingering. If the pair cannot be played comfortably together, use paper. Successive attacks would be a different task and do not count as this simultaneous group.
How many note-attack events, distinct attack times and elapsed seconds does each version have? In V, what changes: root, bass, meter or upper group? To restore R to F, which rest record must change besides adding the bass? Every time is an offset measured in quarter-note beats from the version’s start; 0 is the beginning of beat one, not “beat zero.”
| Role | Pitch / key | Hand / voice |
|---|---|---|
| Bass | F3 / 33 | LH / Bass |
| Inner | A3 / 37 | RH / Inner |
| Upper | C4 / 40 | RH / Upper |
Version onsets
| Version | Meter | Bass attacks | Upper-pair attacks | Total beats | Recovery |
|---|---|---|---|---|---|
| T | 3/4 | 0: F3 | 1, 2 | 6 | 3–6 |
| F | 4/4 | 0: F3, 2: F3 | 1, 3 | 8 | 4–8 |
| V | 4/4 | 0: F3, 2: C3 | 1, 3 | 8 | 4–8 |
| R | 4/4 | none | 1, 3 | 8 | 4–8 |
Check your answers
T has 5 note events, 3 attack times and lasts 6 seconds. F/V each have 6 events, 4 times and last 8 seconds; R has 4 events, 2 times and lasts 8 seconds. Paired notes are simultaneous, so note count cannot determine elapsed duration. V changes only the second bass pitch/key, not planned root F, meter or upper group. R→F restores LH/bass holds at 0–0.5 and 2–2.5, removing those intervals from its full-span rest.
O: if a pair is counted as two beats, circle events sharing an onset. If R still says full rest after restoring bass, repair the timeline before playing. If the score’s voice release is unknown, mark it pending rather than filling it with pedal. Paper checking is independently complete without a listening verdict about style or tone.
Optional practice: alternate attacks/releases and keep empty time
When ready, begin with complete R: two RH groups, LH resting. Count both whole bars without skipping the initial empty beat. Then restore F’s bass as specified. T and V are separate disclosed comparisons, optional rather than a sequence pretending to reproduce a source piece. Keep a comfortable intended level, use no pedal and require no combined hand span. Release each note after half a beat; this timing is not a universal definition of staccato.
Each row is a voice: bass=LH, inner=RH A3, upper=RH C4. B1/B2 label the first/second bass attack; Q1/Q2 the first/second upper group, with i/u for inner/upper voice. Each bracket is one bar; cells are half a beat. An ID starts and holds that note for half a beat; . is a rest in that voice. IDs in the same cell of the two RH rows mean simultaneous attack/release. All three voices rest throughout every version’s final bar.
T bass: [B1 . . . . .] [. . . . . .]
T inner: [. . Q1i . Q2i .] [. . . . . .]
T upper: [. . Q1u . Q2u .] [. . . . . .]
F bass: [B1 . . . B2 . . .] [. . . . . . . .]
F inner: [. . Q1i . . . Q2i .] [. . . . . . . .]
F upper: [. . Q1u . . . Q2u .] [. . . . . . . .]
V bass: [B1 . . . B2 . . .] [. . . . . . . .]
V inner: [. . Q1i . . . Q2i .] [. . . . . . . .]
V upper: [. . Q1u . . . Q2u .] [. . . . . . . .]
R bass: [. . . . . . . .] [. . . . . . . .]
R inner: [. . Q1i . . . Q2i .] [. . . . . . . .]
R upper: [. . Q1u . . . Q2u .] [. . . . . . . .]T holds bass for 0.5 beat and each RH voice for 1 beat. F/V hold bass for 1 beat and each RH voice still for 1; R holds bass for 0. All remaining time is rest in each voice. Recovery is 3–6 in T and 4–8 in F/V/R. T’s last group releases at 2.5; F/V/R at 3.5. No pedal or overlapping holds are specified; acoustic decay is unmeasured.
To restore R→F, keep RH events unchanged and add B1/B2; LH/bass rests change from 0–8 to 0.5–2 and 2.5–8. V→F restores only B2 from C3/key 28 to F3/key 33. T is not F with a note deleted: it specifies 3/4, groups at 1 and 2, and recovery at 3–6. Do not delete a beat from F while retaining its old bar lines.
After a break, hide the tables, reconstruct F’s three voices and explain why V does not automatically change the root: a retention retest. Returning to a real phrase requires restoring its meter, bass, durations and hand assignment before practicing the join. Ability to perform it with one hand needs separate checking. An unseen-score transfer task requires separately prepared, undisclosed material; this package neither provides nor validates it.
Limits and related reading
Source inventory gives 380.054 seconds; the same-hash container gives 380.053. The difference is unexplained, neither is audition/viewing time. The last ASR segment also ends beyond the container duration and is excluded without inferring a cause. Original-audio audition, continuous movement, full source notation, pedal effect, physical suitability, efficacy and application behavior remain unvalidated.
Source scope for this package
Role, event and timeline checks