The moment a song stops being a lyric and becomes a recording is the moment it becomes real. Recording at home is easier than ever — a USB microphone, headphones and a free DAW are genuinely enough to make a demo that sounds like a song. The difference between an amateur-sounding recording and a decent one is almost never the gear; it is four habits: mic placement, gain, one-take discipline, and three mix moves. This guide covers all four.
What you actually need
A USB condenser microphone (any decent brand — around fifty to a hundred dollars), closed-back headphones (to hear yourself without the mic hearing the speakers), and a free DAW: Audacity, GarageBand, Waveform, or the free tier of any major DAW. That is the complete list. Skip the audio interface, the studio monitors, the vocal booth — a closet full of clothes works better than a booth. If you do not even have a microphone, your phone with a wired headset microphone is an acceptable start; record in a small room with soft furniture.
Mic placement: where you stand is 80% of the sound
Three rules, all about physics. 1. Distance. Sing 10–15 centimeters from the capsule — close enough for a warm, present tone, far enough to avoid boominess. If your plosives (p, b) make the meter jump, back off slightly or tilt the mic. 2. Off-axis. Aim the mic slightly above your mouth at a 15° angle rather than straight on; this cuts breath pops without any pop filter. 3. The room. Never face a wall or window — the reflection will smear your vocal. Stand facing into the room, with soft surfaces (bed, curtains, carpet) behind the mic. A small walk-in closet with hanging clothes is genuinely one of the best vocal rooms in any home.
Gain: set it once, then trust it
Open the DAW's input level (gain) control. Sing your loudest line of the song — the chorus peak — and turn the gain until the meter peaks around -12 to -6 dB (yellow zone, never red). That headroom protects you from clipping during loud phrases and keeps the noise floor low. Now do not touch it again during the session. Beginner errors: gain too high (distortion that no plugin can fix) and singing farther or closer as you go (a vocal that changes volume mid-song). Set gain, stand still, sing consistent.
One-take discipline: the best demo habit you will ever learn
Do three full run-throughs of the song with the mic open but not recording — these warm you up and reveal the weak lines. Then hit record and sing the whole song once, top to bottom, mistakes and all. Do not punch-in-fix every phrase; a one-take performance has an energy that comping never captures, and for a demo that energy is the whole point. If a section is truly wrong, record one full second take, not twenty fragments. Then stop: endless takes destroy the performance and teach you to distrust your own voice. The demo's job is to capture the song's potential, not its perfection.
The three mix moves that save any demo
Mix is where beginners drown in plugins. You need exactly three moves: 1. Align and trim. Slide your vocal track so the syllables land on the beat, and trim the head and tail of each clip to kill chair creaks and stray noises. 2. Balance. Set the vocal level so the words are clearly on top of the music, then drop the music a touch further — demos fail by hiding the voice under the track. 3. One gentle compressor, one gentle reverb. A compressor with a 3:1 ratio smooths the loud phrases; a reverb with a short decay adds space without swamp. Then export at -14 LUFS (most DAWs have a loudness normalize) as MP3 320kbps and WAV.
Turning SongSpark into your session player
If you have no instruments, Melody Maker becomes your backing track: generate a melody in your chosen key and scale, download it as MIDI from the tool, and drag the file into your DAW. Assign a piano or pad instrument to the MIDI track, add a simple drum pattern from the DAW's library (or tap your tempo in BPM Tap and set the DAW to match), and you have a chord-and-melody bed to sing over. Structure Planner told you the section order; Chord Mood Map told you the chords feel; now the DAW turns those decisions into audio you can hear while you record.
The finishing checklist before you share
1) Song plays at a steady tempo end to end — 2) vocal is intelligible at a low volume — 3) no clipping anywhere (meters never touch red) — 4) file is exported as both MP3 and WAV — 5) you have listened on your phone speakers AND your headphones. If it passes those five, it is shareable. Your first demo will not be your best recording, and it is not supposed to be. It is the first evidence that the song exists. Share it with one friend you trust, get one honest note, and make the next version better. That loop — write, record, share, improve — is how every recording artist you admire started.
一首歌从一份歌词变成一段录音的那一刻，就是它真正诞生的时刻。如今在家录音比以往任何时候都容易——一支 USB 麦克风、一副耳机和一个免费 DAW，确实足以做出一首听起来像歌的小样。业余录音与像样的录音之间的差别，几乎从来不在设备，而在于四个习惯：麦克风摆放、增益设置、一遍成型的纪律，以及三个混音动作。本指南逐一讲解这四件事。
你真正需要的东西
一支 USB 电容麦克风（任何像样的品牌——大约五十到一百美元）、封闭式耳机（让你听见自己，同时麦克风不会收进扬声器的声音），以及一个免费 DAW：Audacity、GarageBand、Waveform，或任何主流 DAW 的免费版本。这就是完整清单。音频接口、监听音箱、人声隔音棚都可以跳过——一个挂满衣服的衣橱比隔音棚更好用。如果你连麦克风都没有，用手机配有线耳麦的麦克风也是可以接受的起点；在摆有柔软家具的小房间里录制即可。
麦克风摆放：你站的位置决定了 80% 的声音
三条规则，都与物理有关。1. 距离。 在距麦克风振膜 10–15 厘米处演唱——足够近以获得温暖、靠前的音色，又足够远以避免低频轰鸣。如果你的爆破音（p、b）让电平表跳动，就稍微退远一点或把麦克风倾斜。2. 偏离轴线。 让麦克风以 15° 角略微对准你嘴的上方，而不是正对；这样无需防喷罩就能削减气息爆破声。3. 房间。 绝不要面对墙壁或窗户——反射声会让你的人声发糊。面向房间站立，让柔软的表面（床、窗帘、地毯）位于麦克风后方。一个挂满衣物的小型步入式衣橱，确实是家中最好的人声房间之一。
增益：设好一次，然后相信它
打开 DAW 的输入电平（增益，gain）控制。唱出整首歌中最响的一句——副歌的最高点——转动增益，直到电平表峰值在 -12 到 -6 dB 左右（黄色区域，绝不触红）。这样的余量（headroom）能保护你在响亮乐句中不出现削波（clipping），并保持低底噪。之后整个录音过程中都不要再碰它。新手常犯的错误：增益过高（造成任何插件都无法修复的失真），以及演唱时忽远忽近（人声在歌曲中途音量变化）。设好增益、站稳位置、保持一致地唱。
一遍成型的纪律：你将学到的最好小样习惯
打开麦克风但不录音，完整唱三遍——这既能热身，也能暴露薄弱的句子。然后按下录音，把整首歌从头到尾完整唱一遍，连错误一起保留。不要每一句都用补录（punch-in）去修；一遍到底的演唱有一种拼接（comping）永远无法捕捉的能量，而对于小样来说，这种能量就是全部意义。如果某一段确实错了，就完整重录第二遍，而不是录二十个碎片。然后停下来：无止境地重录会毁掉表演，还会让你学会怀疑自己的声音。小样的任务是捕捉歌曲的潜力，而不是它的完美。
能挽救任何小样的三个混音动作
混音是新手最容易淹没在插件中的环节。你只需要三个动作：1. 对齐与修剪。 滑动人声轨道，让音节落在拍点上，并修剪每个片段的开头和结尾，消除椅子吱呀声和杂散噪音。2. 平衡。 设定人声音量，让歌词清楚地压在音乐之上，然后再把音乐往下拉一点——小样失败往往是因为人声被埋在伴奏底下。3. 一个温和的压缩器，一个温和的混响。 3:1 比率的压缩器能平滑响亮的乐句；短衰减（decay）的混响能增加空间感而不浑浊。然后以 -14 LUFS 导出（多数 DAW 都有响度标准化功能），导出 MP3 320kbps 和 WAV 两种格式。
把 SongSpark 变成你的录音乐手
如果你没有乐器，Melody Maker 就能成为你的伴奏：在你选定的调和音阶中生成一段旋律，从工具中下载为 MIDI 文件，拖进你的 DAW。给这条 MIDI 轨指定一个钢琴或铺底（pad）音色，再从 DAW 素材库中加一个简单的鼓点模式（或在 BPM Tap 中敲出你的速度，让 DAW 与之匹配），你就有了一轨可以演唱的和弦加旋律基底。Structure Planner 告诉了你段落顺序，Chord Mood Map 告诉了你和弦的感觉；现在 DAW 把这些决定变成你录音时能听到的音频。
分享前的收尾清单
1) 整首歌从头到尾速度稳定——2) 人声在小音量下依然清晰可辨——3) 没有任何削波（电平表绝不触红）——4) 文件同时导出为 MP3 和 WAV——5) 你既用手机扬声器听过、也用耳机听过。通过这五条，就可以分享了。你的第一个小样不会是你最好的录音，它也不应该是。它是这首歌存在的第一份证据。把它分享给一位你信任的朋友，听取一条诚实的反馈，然后让下一版变得更好。这个循环——写、录、分享、改进——正是你所钦佩的每位录音艺术家起步的方式。