How to Mix Audio for Phone Speakers: The Complete Technical Approach

The general sound design guide covers the audio pipeline for vertical drama as a workflow. This guide covers the physics and psychoacoustics of what happens when that pipeline's output plays through a consumer phone speaker.

They are different subjects. The workflow guide tells you what to do. This guide tells you why the decisions at each stage produce the results they do on the specific acoustic device that 95% of vertical drama viewers are using.

70% of streaming happens on phones. If the vocal disappears on a phone speaker, the mid-range is wrong. That is the complete diagnostic for the most common audio failure in vertical drama post-production stated in two sentences. Most productions discover this diagnostic in the acquisition rejection note rather than during the mix.

The phone speaker is not a smaller version of studio monitors. It is a fundamentally different acoustic device with different frequency response characteristics, different power handling, different dispersion patterns, and different psychoacoustic interactions with ambient noise. Every audio decision in vertical drama post-production produces a different result through a phone speaker than through studio monitors, and understanding specifically what changes and why is what separates a mix that passes the phone playback test from one that does not.

The Phone Speaker: What It Actually Is

A phone speaker in 2026 is a miniature transducer, typically 10 to 15mm in diameter, driving a small enclosure with minimal acoustic reinforcement. The fundamental physical constraints of this configuration determine its frequency response characteristics before any signal processing is considered.

The frequency response chart of a phone speaker shows deviations from the ideal flat line. Any peaks or valleys are audible distortions of the sound. An ideal speaker produces a flat response across all frequencies. No phone speaker produces a flat response. Every phone speaker has a characteristic response curve that the audio engineer mixing for phone delivery has to understand and account for.

Low-frequency cutoff. Consumer phone speakers cannot reproduce frequencies below approximately 150 to 250 Hz with meaningful acoustic output. The specific cutoff varies by phone model and speaker design. Flagship phones with dual speaker systems and computational audio processing can extend meaningful bass response to 80 to 100 Hz, but the majority of mid-range phones that represent the bulk of the vertical drama viewing audience have effective bass response that begins to roll off sharply below 200 Hz and produces minimal acoustic energy below 150 Hz.

The practical implication: any audio information below 150 Hz in a vertical drama mix is inaudible to the majority of the audience. It is not reproduced by the speaker. It consumes headroom in the mix without contributing to the listener's experience.

Mid-range presence peak. Most phone speakers have a presence peak in the 1,000 to 4,000 Hz range where they are most acoustically efficient. This is the frequency range where consumer speakers are physically best suited to produce output, and it is also the frequency range where dialogue intelligibility lives. Voice consonants, which carry the linguistic content of dialogue, occupy the 1,000 to 8,000 Hz range. The phone speaker's presence peak in this range means that dialogue is not the problem area for phone speaker mixing. The problem areas are the frequency ranges above and below the presence peak where the speaker's efficiency drops.

High-frequency rolloff. Phone speakers roll off above approximately 10,000 to 12,000 Hz. The air and detail that studio monitors reproduce in the 12,000 to 20,000 Hz range are absent from phone speaker output. A mix that depends on high-frequency content for its sense of space, depth, or sheen will lose those qualities entirely on phone speakers.

The LUFS Target and Why It Differs From Broadcast

The most common technical audio error in vertical drama post-production is mixing to broadcast loudness standards rather than mobile streaming standards. The gap between these two standards is not marginal. It is a 9-decibel difference in integrated loudness that produces audibly different results on a phone speaker in ambient noise.

Most streaming platforms normalize to -14 LUFS integrated. Upload at that target and the platform applies minimal processing. Upload quieter and you force viewers to adjust volume manually, which leads to lower engagement and more drop-offs.

Broadcast television standards in the EU require -23 LUFS integrated under EBU R128. US broadcast requires -24 LUFS integrated under ATSC A/85. These standards were developed for the living room listening environment: a television speaker or soundbar in a quiet room, at a comfortable listening distance, with the viewer's full attention.

The vertical drama viewer is not in that environment. They are in ambient noise, holding a phone at arm's length, often without headphones, with the phone speaker competing against the acoustic environment around them. A mix at -23 LUFS broadcast standard produces audio that requires the viewer to turn the phone to maximum volume to hear dialogue clearly in any ambient noise condition. Maximum phone volume introduces its own distortion characteristics from the speaker's power handling limits.

A mix at -14 LUFS produces audio that is intelligible at a moderate phone volume setting in typical ambient conditions without the viewer needing to adjust volume, and without pushing the phone speaker's power handling to its distortion threshold.

The LUFS targets that matter for vertical drama delivery:

Integrated loudness: -14 LUFS. The target for TikTok, YouTube, and the vertical drama platforms that specify mobile streaming standards. This is the loudness level at which the mix competes effectively with ambient noise without requiring the viewer to reach for the volume control.

True peak: -1 dBTP maximum. True peak is not the same as sample peak. Inter-sample peaks occur during digital-to-analog conversion when the curve connecting digital samples can overshoot the 0 dB digital ceiling. When platforms transcode the delivery file to AAC or another lossy format, the encoding math adds 0.3 to 0.5 dB of gain. A mix with true peaks at 0.0 dB will clip after platform transcoding. -1 dBTP provides the headroom that prevents post-transcoding clipping.

Loudness range: 4 to 8 LU. Loudness range describes the statistical spread between the quieter and louder sections of the mix. A loudness range below 4 LU indicates a heavily compressed mix with no dynamic range: dense and fatiguing to listen to across a 90-second episode. A loudness range above 10 LU indicates a mix with dynamics that are too wide for phone speaker playback in ambient noise: the quiet sections become inaudible and the loud sections produce distortion at the speaker's power handling limit. The 4 to 8 LU range balances dynamic expressiveness with consistent phone speaker intelligibility.

Bass-Rolling for Phone Playback

Bass-rolling is the most specific technical adjustment that phone speaker mixing requires and the one most frequently omitted by engineers trained on conventional broadcast and music mixing.

Bass-rolling applies a high-pass filter to the mix's low-frequency content, rolling off the frequencies below the phone speaker's effective response range. The frequencies below 150 to 200 Hz that the phone speaker cannot reproduce do not disappear from the mix when the phone speaker fails to reproduce them. They continue to exist in the audio signal, consuming headroom, producing inter-modulation distortion as the speaker attempts to move the driver against physical limits it cannot satisfy, and muddying the mid-range frequencies where the phone speaker has its best performance.

The specific bass-roll approach for vertical drama phone speaker mixing:

The dialogue track high-pass filter. Apply a high-pass filter to the dialogue track at 80 to 100 Hz with a 12 dB per octave slope. This removes the subsonic and low-bass content from dialogue recordings that adds no intelligibility value and consumes headroom on the phone speaker. A dialogue track that has been high-pass filtered at 80 Hz sounds identical on studio monitors and sounds significantly cleaner on phone speakers because the low-frequency content that the phone speaker was struggling with has been removed.

The music track high-pass filter. Apply a high-pass filter to the music bed at 150 to 200 Hz with a 6 dB per octave slope. The music's bass content that lives below 150 Hz is inaudible on the phone speaker and occupies headroom that the mid-range music content needs to register clearly. A music bed mixed with the sub-150 Hz content present sounds fuller on studio monitors and muddier on phone speakers than the same music bed with the sub-150 Hz content removed.

The effects track high-pass filter. Sound effects with significant low-frequency content, a door slam, a footstep on a hard surface, an atmospheric drone, can produce clipping or mechanical distortion from the phone speaker's driver when played at -14 LUFS integrated. Applying a high-pass filter to effects at 100 to 120 Hz prevents the effects' low-frequency content from pushing the phone speaker beyond its mechanical limits.

The bass-roll does not make the mix thin on phone speakers. It makes the mix clearer on phone speakers by removing content the speaker cannot reproduce and which was interfering with the content it can reproduce. The monitor-based listening impression of the bass-rolled mix is a slightly thinner low end, which is the correct impression because the phone speaker's response below 150 Hz is effectively zero.

Dialogue Clarity Under Ambient Noise Competition

The ambient noise competition problem is the most commercially significant audio challenge in vertical drama mixing and the one that requires the most specific technical response.

Ambient noise occupies the frequency range below 1,000 Hz primarily. HVAC noise, traffic, cafe background sound, and public transport noise all have energy concentrated in the low and low-mid frequency range. The dialogue that competes against this ambient noise for the viewer's attention occupies the 500 Hz to 4,000 Hz range primarily, with the consonants that carry linguistic content concentrated in the 1,000 to 8,000 Hz range.

A mix where dialogue sits at its natural tonal balance relative to the music and effects bed competes effectively against ambient noise at close listening distance on studio monitors. The same mix competes poorly against ambient noise on a phone speaker at arm's length in a noisy environment because the phone speaker's limited power handling cannot achieve the sound pressure level at close range that studio monitors achieve with much more acoustic energy.

The specific adjustments that improve dialogue clarity under ambient noise competition:

Mid-range presence boost on dialogue. A shelf or bell boost of 2 to 4 dB in the 2,000 to 4,000 Hz range applied to the dialogue track increases the perceptual presence of consonant sounds in the viewer's ambient environment. The 2,000 to 4,000 Hz range sits above the ambient noise's primary energy concentration and within the phone speaker's efficient reproduction range. Boosting this range makes dialogue cut through ambient noise without increasing the overall dialogue level.

Mid-range notch on music under dialogue. Rather than simply reducing music level under dialogue, a notch of 2 to 3 dB in the 1,000 to 3,000 Hz range applied to the music track wherever dialogue is present reduces the music's competition with dialogue in the frequency range where dialogue intelligibility lives. This approach is more frequency-specific than a broadband level reduction and produces better dialogue separation without audibly reducing the music's presence at other frequencies.

Dialogue compression for consistent level. A vertical drama production shooting 15 to 20 pages per day produces dialogue that ranges from whispered intensity to projected confrontation within the same episode. The dialogue compression stage in post-production normalizes this dynamic range so that quieter delivery remains audible on the phone speaker in ambient noise and louder delivery does not produce clipping. The target for dialogue level consistency is a momentary LUFS range of approximately 4 to 6 LU across the full episode. A performance that ranges more widely than this requires either level riding or a more aggressive compressor than the dialogue compression would normally require.

Speech intelligibility test. The phone intelligibility test for dialogue is specific: listen to the episode on the phone speaker without watching the screen, in a room with a television or radio playing at moderate level in the background. Can every word of dialogue be understood without straining? If any line requires effort to understand in this condition, the dialogue level, presence boost, or music-under-dialogue notch requires adjustment.

The Stereo-to-Mono Consideration

Most phone speakers are mono. Premium phones with dual speaker systems produce stereo separation that is perceptually narrow at phone viewing distance. A mix that depends on stereo placement for its spatial clarity produces a different and less spatially clear result on a phone than on studio monitors.

Stereo placement decisions that work on studio monitors and on headphones do not carry across to phone speakers because the physical separation between a dual phone speaker system is 5 to 10 cm at maximum, and at arm's length viewing distance this produces minimal stereo field rather than the wide stereo stage that studio monitors produce.

The practical adjustments for stereo-to-mono translation:

Check mono compatibility before finalizing the mix. Route the stereo mix to a mono bus and listen on the phone speaker. Does the dialogue hold its level and clarity in mono? Does the music retain its tonal character in mono? Does any stereo element produce phase cancellation in mono that causes it to drop in level or change in character? Phase cancellation in mono is the most common stereo mixing problem that surfaces in phone speaker playback and is invisible in studio monitor listening.

Center the dialogue in mono. Dialogue should be centered in the stereo field regardless of the character's physical position in the scene. A character staged camera-left in the shot who has their dialogue panned camera-left in the mix produces a mono collapse of the dialogue toward center that shifts the perceived dialogue level relative to the music. Dialogue centered in the stereo field produces consistent level in both stereo and mono playback.

Reserve the stereo field for music and effects. Music and effects can use the stereo field freely as long as they pass the mono compatibility check. The drama of a wide stereo soundscape adds nothing to phone speaker playback but adds to headphone playback, which is a secondary delivery environment for vertical drama. The stereo field is available as an enhancement for the viewer who uses headphones, not as a design dependency for the viewer who does not.

The Batch Mastering Workflow for 70 Episodes

A properly configured batch mastering workflow processes 50 episodes in approximately 30 to 45 minutes, including loudness measurement, limiting, and format conversion.

The batch mastering workflow for vertical drama audio delivery:

The mastering chain is configured as a template that applies consistently to every episode in the batch. The chain contains: a high-pass filter at 80 Hz removing subsonic content from the full mix, a multiband compressor addressing the 200 to 500 Hz range where phone speakers tend to build up muddiness from overlapping dialogue, music, and effects content, a limiter with output ceiling at -1 dBTP, and a loudness normalizer targeting -14 LUFS integrated.

The template is applied to every episode in the batch without episode-specific adjustment. The only episode-specific review is a loudness measurement confirmation after the template is applied: does the episode's integrated LUFS reading fall within ±1 LU of the -14 target? Episodes outside this range require individual review to identify whether the source mix has a level issue before the mastering template was applied.

The phone playback spot-check applies to at minimum episodes 1, 5, 10, 25, and 50 in any batch. These episodes are played on the phone speaker in ambient conditions, not through the mastering monitor chain. If any spot-check episode fails the phone intelligibility test, the mastering template requires adjustment before the batch is finalized.

Axis AI Studios Perspective

The audio failure that produces the most acquisition rejections in vertical drama post-production is not a technical error that appeared during the mix. It is a reference environment error: mixing was done correctly for the reference environment but the reference environment was not the phone speaker.

A mixer who builds their dialogue clarity decisions against studio monitors, who accepts bass extension that produces monitor-appropriate fullness, and who approves a loudness level calibrated for broadcast delivery has not made audio errors. They have made correct decisions for an incorrect reference environment.

The adjustment this requires is not better mixing technique. It is a different sequence: phone playback testing at every mix approval stage, not as a final delivery check, and bass-rolling and loudness decisions made specifically for the phone speaker environment rather than optimized for monitors and corrected for phone afterward.

At Axis AI Studios, the phone speaker test is the primary approval criterion for every audio mix, at every stage of the pipeline. The monitor is a tool for technical decision-making. The phone is the reference for commercial decision-making. The two environments require the same mix to produce different results, and building the mix to pass both is the technical discipline that vertical drama audio requires.

For production companies who want to commission vertical drama from a post-production partner whose audio workflow is built around the actual delivery environment, reach out at business@axisaistudios.com.

FAQ

Why Is -14 LUFS the Target for Vertical Drama and Not the Broadcast Standard of -23 LUFS?

The broadcast standard of -23 LUFS was developed for the living room listening environment: television speakers or soundbars in quiet rooms at comfortable listening distances. Vertical drama is consumed on phone speakers in ambient noise at arm's length. A mix at -23 LUFS requires the viewer to turn the phone to maximum volume to achieve dialogue clarity in any ambient noise condition, which introduces distortion from the phone speaker's power handling limits. A mix at -14 LUFS achieves dialogue clarity at moderate phone volume in typical ambient conditions without approaching the speaker's power handling limits.

Does Bass-Rolling Make the Mix Sound Thin on Studio Monitors?

Yes, slightly. A mix that has been bass-rolled for phone speaker delivery sounds slightly thinner below 150 Hz on studio monitors than a mix without the bass-roll. This is the correct perceptual outcome because the phone speaker's effective frequency response begins at approximately 150 to 200 Hz. The bass content below that frequency that sounds full on studio monitors is not reproduced by the phone speaker. Removing it from the mix does not reduce the listening experience on the phone. It removes content that was consuming headroom without contributing to what the viewer hears.

How Do You Test a Vertical Drama Mix for Phone Speaker Clarity Before Delivery?

Play the completed episode on two consumer phone models, one flagship and one mid-range, without headphones, in a room with background noise at the level the viewer is likely to use. The background noise level should approximate a domestic environment: a television at moderate volume or a radio playing in the background. Listen to the episode without watching the screen. If every word of dialogue is intelligible without effort in this condition, the mix passes the phone clarity test. If any line requires concentration to understand, the dialogue presence boost, music-under-dialogue notch, or dialogue compression requires adjustment before delivery.

Further Reading

For the sound design pipeline that the phone speaker mixing decisions described in this post complete, the sound design guide for vertical micro-dramas covers the complete audio approach from production recording through mobile-calibrated delivery.

For the voice acting and ADR decisions that determine what raw material the phone speaker mix has to work with, the voice acting and ADR guide for vertical micro-dramas covers performance direction, session management, and sync precision in detail.

For how the audio mix interacts with the color grade and VFX pipeline in the complete post-production workflow, the vertical drama post-production guide covers the full pipeline from rough cut through delivery specifications.