Production of a series of 10 audio clips for a pharmacy chain - recording, sound design, loudness mixing and technical verification

Initial situation: the pharmacy chain has 120 points across the Russian Federation, each with 100V ceiling speakers and a media player that plays advertising in the hall. Now there are scattered videos of different volumes, some of them “hiss” and are difficult to understand against the background noise of the hall. It is necessary to bring everything to a single sound and prepare a new package of audio clips for the specific acoustics of pharmacies.

What needs to be done: produce 10 advertising and informational audio clips, each lasting 15-20 seconds, for internal broadcasts in a pharmacy. This is not a podcast or an announcer’s reading “in a vacuum” - full-fledged work with sound is required so that speech remains intelligible at a real point with a limited frequency band and background noise.

Technical environment: playback via 100V linear ceiling speakers, media player accepts WAV 48 kHz 16-bit PCM and MP3 CBR 320 kbps. In the hall there is constant noise from refrigerators and conversations, and the resonances of the room can often be heard. I’ll give you 2 recordings of a test from a typical pharmacy (30-60 seconds each) and 3 old videos as an example of problems.

Stages and scope of work:
1) Pre-production - analysis of test recordings, selection of a technical processing profile (EQ/dynamics/de-esser/saturation) for “pharmacy” acoustics and formation of a unified sound of the series.
2) Voice recording - 1 main voice (Russian, neutral, 30-45 years old), recording in a controlled environment, without audible noise/room. If you work with your studio and actor - ok, if you do the voice acting yourself and can meet the requirements - also ok.
3) Sound design and editing - neat backgrounds and accents without conflicts with intelligibility, equalization of tempo and pauses, cleaning of breath/clicks, control of sibilants.
4) Mixing and mastering - uniform volume and tonal balance between all 10 videos, preparation in 2 formats.
5) Technical check - emulation of a limited band and listening on a “bad” speaker, control of articulation and readability against the background noise (I’ll give you a background loop from the pharmacy).

Result: 10 final videos + 10 text timecode sheets (where the announcer's phrase begins and ends, without complex markings). Separately, you need the project sources (DAW session) or themes (voice/music/sfx) - your choice, but so that you can replace 1 phrase in the future without rebuilding from scratch.

Restrictions:
- The duration of each video is strictly 15-20 seconds, without “tails” of silence longer than 200 ms.
- Videos should sound correct and without clipping after transcoding to MP3.
- Do not use recognizable/protected brands and samples without a license. Music and SFX are either yours licensed for commercial use or completely original.
- Speech should remain intelligible when listening on a smartphone in speaker mode and in mono.

Measurable acceptance criteria:
- Loudness: -16 LUFS integrated for the MP3 version and -18 LUFS integrated for the WAV version, true peak not higher than -1.0 dBTP.
- Volume variation between videos is no more than 1 LU (integrated).
- Noise/background on the voice track during pauses - below -60 dBFS (with normal gain, without aggressive gating, which eats up the endings of words).
- No audible clicks, digital artifacts, overload, or “sand” on sibilants.

Materials from me: texts of 10 videos, a list of restrictions on wording (pharmacy), 2 test recordings from a pharmacy, background loop, examples of old videos.

Tools: I expect to work in a professional DAW (Pro Tools/Reaper/Logic, etc.), loudness measurement (EBU R128), true peak control, voice processing (de-esser, dynamics, EQ) and preparation of WAV+MP3 formats.

Deadlines: it is advisable to keep it within 10-14 calendar days, with 2 rounds of text/intonation edits (not a rework of the concept, but targeted adjustments).