Best practice mixing for spoken-word interviews focuses on maximizing speech intelligibility, maintaining a consistent volume to avoid “volume surfing,” and ensuring high-quality, natural sound across different listening environments.
1. Record voices in mono
All speech is treated as mono and centred.
Typical setup:
Presenter mic → mono → centre
Guest mic → mono → centre
Remote contributors → mono → centre
Even if you record two microphones, they are usually kept as separate mono tracks but mixed centre.
This ensures:
compatibility with mono radios
consistent voice level
no listener distraction.
2. Use stereo only for atmosphere and music
Stereo is mainly reserved for:
music beds
sound effects
location atmos (crowds, traffic, etc.)
Speech remains centred while stereo elements create the sense of space around it.
Example:
Voice: centre
Background ambience: stereo
Music bed: stereo but lower level.
3. Voice levels
BBC radio speech typically sits around:
Peak: about -6 dBFS
Average: about -16 to -18 LUFS for speech segments
This keeps speech clear without clipping once it goes through broadcast processing.
4. Consistent mic tone
Broadcasters try to keep presenter and guest sounding similar in tone and loudness.
Typical techniques:
gentle compression (around 3:1 ratio)
light EQ to remove rumble (HPF around 80–100 Hz)
slight presence boost if needed around 3–5 kHz.
5. Avoid extreme stereo imaging
For spoken-word programmes:
no hard L/R panning of voices
no stereo widening on dialogue
avoid phase-heavy processing.
This keeps everything robust when summed to mono, which still happens on some transmission chains.
6. Mix for “small speaker clarity”
BBC radio mixes are tested so they work on:
car radios
smart speakers
phones
kitchen radios
If speech is clear on a single small speaker, the mix is considered good.
7. Music bed balance
Typical spoken-word balance:
Voice: main level
Music bed: -20 to -25 dB under speech
The bed should support, not compete with speech.