The Audio Production Hub for Malaysian Recording and Podcasting

Record clean speech first. Everything else fails without it. If you cannot capture a voice free of room noise, no loudness trick, plugin, or editing sequence will rescue the file. This guide routes you through four real problems: recording clean speech, staging gain, editing spoken audio, and hitting a loudness target that survives on any platform. Start with the problem in front of you.

If your gain staging is wrong, your edits will fight you. If your editing sequence is backwards, you will redo work. And if you do not know why a radio station and a streaming platform want different loudness, you will chase a number that does not apply to your release. The contract here is simple: explain signal flow and trade-offs accurately, avoid universal gear recommendations, and never promise audience growth or income. What you build is your own.

• First decision: What are you making? A podcast episode, a speech recording for a client, a radio-ready voiceover, or a piece of music with vocals?
• First skill: Gain staging. Everything else depends on a clean signal before you touch a single plugin.
• Most common failure: Recording too hot, then trying to fix distortion in editing. Distortion cannot be fixed.
• What the BFM 89.9 production standard actually is: A working reference for clean, intelligible spoken audio, not a legal requirement.
• What the RTM broadcast standard actually is: A technical and regulatory benchmark for broadcast, not a podcasting target.
• Platform variation: Spotify, Apple Podcasts, YouTube, and radio all have different loudness targets. Aim for the platform you are actually publishing on.

Pick the route that matches the work in front of you. Each route below links to a dedicated guide that owns the detail.

Route one is clean speech. If your recording has room noise, mouth clicks, or an uneven tone, stop and fix the recording environment before you touch anything else. Route two is gain staging. If you do not understand headroom, you will either record too quiet and amplify noise, or too hot and clip. Route three is editing spoken audio. The sequence matters: cut first, then process, then balance. Route four is loudness for podcasts. If you are publishing to Spotify or Apple Podcasts, you need a LUFS target. If you are submitting to radio, you need a different one.

Do not treat these as sequential stages. They are parallel skills. A podcast producer who records clean but edits badly still ships bad audio. A voiceover artist who stages gain correctly but ignores loudness measurement will get rejected by a client. The skill is knowing which route to walk at each step.

The table below compares the four production priorities against five axes that decide whether your audio works: typical failure, fixable in post, primary tool, measurement unit, and the Malaysian reference point. Read it in that order, not the order the columns print. The failure tells you what is wrong. The fixable column tells you whether to re-record or edit. The tool tells you where the work happens. The unit tells you what number to watch. The reference point tells you who in Malaysia already works this way.

Ignore any spec sheet that leads with a plugin brand or a microphone model. That is advertising. The evidence is in the signal flow: did the recording happen in a controlled room, with a microphone technique that placed the capsule correctly, and with gain staging that left headroom? Everything else is a decision you can change later.

Production priority	Typical failure	Fixable in post?	Primary tool	Measurement unit	Malaysian reference point
Clean speech recording	Room reflections, background noise, mouth clicks	Partially. De-noise and de-click help, but a bad room stays bad.	Microphone technique and room acoustics	Signal-to-noise ratio	BFM 89.9 production standard for spoken audio
Gain staging	Clipping at the converter, or recording so low that noise dominates	No. Distortion is permanent; noise amplified is still noise.	Preamp gain and fader discipline	dBFS with headroom, usually -18 to -12 dBFS average	RTM broadcast standard for signal integrity
Editing spoken audio	Cutting words out of context, leaving breaths and room tone mismatched	Yes, if the recording was clean and staged properly.	Editing sequence: cut, then process, then balance	Time and fades	Podcast production workflows, especially talk and true-crime formats
Podcast loudness	Mastering to radio levels, then being too loud or too quiet on streaming platforms	Yes. Loudness is a final-stage correction.	Loudness measurement and a true peak limiter	LUFS integrated and true peak in dBTP	Platform variation in loudness targets, not a single broadcast figure

Podcast production in Malaysia has settled into a distinct pattern. The local market skews hard toward talk, business, and true-crime conversation formats. Audio drama and music shows are rare. That shapes every production decision: you are recording one or two voices in a room, not a cast, and your primary challenge is clean speech over long sessions, not sound design.

Room Acoustics And Microphone Technique

Start with room acoustics. A Malaysian apartment with tile floors and bare walls is a reverb chamber. You do not need foam panels on every surface. You need a reflection point behind the speaker and a reflection point behind the mic, treated with anything from a heavy curtain to a bookshelf. The BFM 89.9 production standard is not a legal requirement, but it is a useful benchmark: their spoken audio is intelligible on a phone in a car. That is the bar. If your recording cannot clear it, re-record.

Microphone technique matters more than the microphone. Keep the capsule a fist-width from the mouth, slightly off-axis to reduce plosives. Do not cup the mic. Do not record in a kitchen. Do not record next to an air conditioner you plan to de-noise later. De-noise is a repair tool, not a production strategy.

What To Skip In Podcast Production

Skip the compressor on the way in. Record clean, at a conservative level, and compress in the editing stage. Skip the idea that you need a broadcast chain. You are not broadcasting. You are publishing a file. Skip the promise that a better microphone will fix a bad room. It will not.

Speech recording in Malaysia is the foundation skill for podcasting, voiceover, corporate narration, and audiobook work. The same principles apply, but the failure modes change with the context. A corporate training video recorded in an untreated meeting room will have a different problem from a podcast recorded at a desk. The meeting room has parallel walls and a table reflection. The desk has a monitor hum and a keyboard.

Choose And Control The Recording Environment

The first decision is the recording environment. For clean speech, you need a room where you can control the sound source, the reflections, and the noise floor. A walk-in wardrobe full of clothes is a better vocal booth than a living room with a tile floor. If you record in a room with a window that faces a main road, schedule the session when traffic is low or move the mic away from the window.

The second decision is the signal flow. Microphone into interface, interface into recording software, recording software set to capture a dry, unprocessed signal at a level with headroom. Do not add reverb, compression, or EQ on the way in. Those are editing decisions. If you print them to the file, you cannot undo them.

Language And Consonant Pressure

Speech recording in Malaysia is also a language question. Malay, English, Mandarin, and Tamil speech have different consonant pressures, sibilance, and plosive energy. A Mandarin recording needs more attention to sibilance. A Malay recording needs consistent vowel placement. An English recording for a corporate client needs a neutral, legible accent. None of these are better or worse; they are distinct problems to solve with the same tools.

Gain staging is the most misunderstood part of audio production in Malaysia, mostly because it is invisible. You cannot hear a good gain stage. You only hear a bad one: distortion, noise, or a recording that falls apart the moment you add processing. The rule is simple: record at a level that leaves headroom between your loudest moment and digital zero. For spoken audio, aim for an average level around -18 dBFS with peaks no higher than -10 dBFS. That gives you room to edit, process, and master without clipping.

The trade-off is real. Record too low and you amplify the noise floor of the room, the mic, and the preamp. Record too high and one emphatic syllable clips the converter, and that distortion is permanent. The sweet spot is not a single number; it is a range that depends on the dynamic range of the voice. A loud, expressive speaker needs more headroom than a quiet, even one.

Do not confuse gain staging with loudness. Gain staging happens at the recording stage, before the file is written. Loudness measurement happens at the mastering stage, after the edit is complete. If you try to fix loudness with gain, you will clip. If you try to fix gain staging with a limiter, you will lock in the noise.

Loudness for podcasts is where most Malaysian producers go wrong, because they carry radio habits into a platform world. Radio has a broadcast standard: the RTM broadcast standard and commercial FM stations process audio heavily to stay competitive in the loudness wars. Podcasts do not. Spotify normalises to about -14 LUFS integrated. Apple Podcasts targets around -16 LUFS. YouTube sits near -14 LUFS. If you master a podcast to -9 LUFS because it sounds loud and proud, the platform will turn it down, and your true peaks will still be too hot for the codec.

LUFS Versus Peak Level

Loudness measurement is not the same as peak level. A file can peak at -1 dBTP and still measure -20 LUFS integrated if the voice is sparse and the silences are long. A file can measure -12 LUFS and peak at -3 dBTP if the mastering chain is disciplined. The number that matters for podcast loudness is integrated LUFS, measured across the whole episode, not the loudest moment.

Calibrate Your Monitoring

Set your monitoring setup so you can hear what you are measuring. If your monitors are too loud, you will mix too quiet. If they are too quiet, you will mix too hot. Calibrate your listening level once, mark the volume knob, and stop touching it. The platform variation is predictable. Your monitoring should be too.

Signal flow is the order in which audio moves from the mic to the final file. Most mistakes are signal flow mistakes. A compressor before a gate sounds different from a gate before a compressor. A de-esser after an EQ behaves differently from a de-esser before it. The rule for spoken audio is: clean first, then edit, then process, then balance. Cut the bad takes. Remove the clicks. Then apply the EQ. Then the compression. Then the loudness normalisation. If you process before you cut, you are processing audio that will be deleted.

The editing sequence matters more than the tools. Cut for content first: remove false starts, long pauses, and mistakes. Then edit for sound: trim breaths, smooth room tone, match levels between speakers. Then process: EQ for intelligibility, compression for consistency. Then measure: loudness measurement, true peak, and platform target. If you do it out of order, you will redo work.

Every decision is a trade-off. More compression means a more consistent level but a less natural voice. More EQ means more intelligibility but a less honest recording. More de-noising means a cleaner background but a more processed, sometimes metallic, tone. The skill is knowing which trade-off serves the material. A true-crime podcast can survive heavy processing. A poetry reading cannot.

Do not buy anything. Do not download another plugin. Record thirty seconds of your voice in the room where you plan to work, using the mic you already have. Set the gain so your average level sits around -18 dBFS. Listen back on headphones, not speakers. If you hear room noise, fix the room. If you hear distortion, lower the gain. If you hear a clean, clear voice with headroom to spare, you are ready to edit. Every other decision follows from that one recording.

This is the single most practical thing you can do next. It costs nothing, takes five minutes, and tells you more about your audio production situation than any spec sheet or forum thread ever will. Do it before you read another word about microphones. Do it before you plan your first episode. Do it now.