Can learned voice cues improve
speech recognition in noise?

1. Question and prior evidence

Which spectral and temporal features of a learned talker improve recognition when carried into new speech? Do their contributions combine additively or interact? Can that profile identify training targets that produce retained gains for previously unheard talkers?

Familiar-talker benefits have been demonstrated in older listeners with hearing loss.¹ Laboratory familiarization can also improve intelligibility and reduce listening effort.² These findings motivate learning as an intervention target; they do not establish which acoustic components produce the benefit.

Speech-cue weighting varies across listeners with hearing loss,³ and changing masker modulation shifts sentence-level spectral weights.⁴ A listener's profile is therefore conditional on the task and acoustic scene. The objective is better recognition, not equal spectral and temporal weights.

2. From music to speech

Originating in musical source selection, the proposal asks whether learned acoustic structure helps listeners follow a target amid competition. Speech makes that benefit directly measurable as word recognition; transfer from music training is not assumed.

3. Testable hypotheses

H1 · Learned component use. Familiarization changes the contribution of at least one prespecified spectral or timing contrast beyond matched exposure to a control voice. Both adding a B-like component to A-like speech and replacing it in B-like speech are tested.

H2 · Joint dependence. Familiarization changes the spectral × timing interaction. Its sign is not prescribed; a null result is informative only if uncertainty excludes a prespecified meaningful interaction.

H3 · Retained transfer. A frozen profile-guided selection policy improves speech reception threshold for new natural talkers four weeks after training, relative to a dose-matched active policy.

4. Research approach

Spectral-weighting methods provide a direct measurement precedent.⁴ A rhythm/EEG preprint links target/background timing to recognition and neural tracking.⁵ The proposed bridge is to ask whether learned component use predicts trainability and retained transfer.

The project combines auditory assessment, learning experiments and computational modelling to relate individual cue use to speech recognition.

ElevenLabs will generate matched-text candidates. R³/T³ will measure the resulting audio, preserving the complete feature vectors and extraction settings. Two recordings with the same plotted distance may differ in their component measurements; X/Y proximity is therefore insufficient for experimental acceptance.

The stimulus pilot will select a small set of spectral and timing contrasts. Each comparison must achieve its intended feature change while bounding off-target changes. The proposed four-cell design proceeds only if enough recordings satisfy these rules across sentences and talker pairs.

5.1 Specify features before testing recognition

R³ contributes envelope, contrast, flatness, flux and modulation channels. These are computational descriptors, not direct formant or phoneme measurements. A proposed speech layer adds validated F0, formant, aperiodic-energy and alignment measurements. Preserve durations on the original clock when aligning spectral comparisons. Preregister the target coordinates and nuisance set; if several terms remain coupled, interpret their joint contrast.

5.2 Qualify comparisons, not just recordings

Retain per-feature target deviations, nuisance deviations and missing-value flags relative to both references. Fix scaling and acceptance tolerances in the independent pilot; audit residual covariance and all four cells across texts. Recheck after playback or hearing-aid processing.⁷ Failed candidates return to generation or resynthesis. Passing the audit supports a specified acoustic comparison; it does not establish perceptual independence.

Stage I will recruit English-speaking adults with bilateral sensorineural hearing loss and an age-matched normal-hearing comparison group. Audiometry, speech-in-noise performance, device use and language background will characterize the sample. Cochlear implant users will not be pooled with acoustic-hearing participants.

Talkers are counterbalanced across roles. Participants are randomized to familiarization with B or matched exposure to a third talker C, absent from the A/B test bank. A is an untrained reference, not a zero-feature voice. New sentences are used in each pre-training, post-training and replication session.

6.1 Mapping in competing sound

Primary mapping uses two competing talkers. A separately analysed speech-shaped-noise condition tests masker dependence. Set an individual challenge level before familiarization, then hold it fixed across that listener's matched cells. Target cues contain different words from the scored sentence. Counterbalance masker identities and trial order.

The listener types the target words. Primary scoring is open-set word accuracy; intrusions, voice identification and reported effort are secondary. Store raw responses and prespecify spelling normalization. Word errors are not automatically attributed to a phoneme or cue.

6.2 From a profile to a training target

Re-estimate spectral, temporal and joint effects on a separate day and new utterances. Then use independent, counterbalanced learning probes to test whether each candidate contribution can change. A small baseline effect is insufficient grounds for targeting it.

The policy uses repeatability, uncertainty and probe response. Thresholds and eligibility rules are fixed in Stage I; report the eligible proportion. Any benefit applies to this defined population of eligible profiles.

6.3 Independent policy trial

In a new SNHL cohort, eligible participants are randomized 1:1 to profile-guided or balanced control selection. Both arms receive the same task families, talker pool, feedback, duration and masker progression. The intended dose is 20 minutes/day, five days/week for four weeks, subject to feasibility confirmation.

Begin near the learned reference and introduce accepted recordings with reduced familiar support. Guidance may prioritize spectral, temporal or joint variation according to the locked rule. The control uses a prespecified balanced schedule; the manipulated factor is selection policy.

6.4 Outcome and retention

The primary endpoint is baseline-adjusted speech reception threshold (SRT) at week 8, four weeks after training ends. Test natural talkers and sentences never used in profiling or training; lower SRT is better. Assessors are blinded. Immediate post-training performance is secondary.

The goal is retained speech recognition in noise. Better recognition does not demonstrate recovery of cochlear sensitivity or normalization of a clinical audiogram.

7. Learning-specific contrasts

Fit word-correct responses with a hierarchical logistic model including component cell, exposure arm, session and hearing group. Model listener, sentence, talker-pair and masker variation. Let q_st denote estimated log odds for spectral state s and timing state t:

The learning-specific effect is the pre-to-post change in C_S, C_T or I after B training minus that after C training. Report probability-scale marginal effects as well. Retain the prespecified feature vectors in the model; report bundle effects where terms cannot be separated. These contrasts measure dependence on the accepted stimulus manipulation; they are not percentages of attention.

8. Clinical and computational tests

For H3, estimate the baseline-adjusted between-arm SRT difference at retention, using intention-to-treat analysis and prespecified missing-data sensitivity analyses. Improved trained-item performance alone is insufficient.

For MI, add T³ summaries to a matched acoustic/exposure baseline. Candidate native spans of 200, 600 and 2,000 ms are modelling choices, not attention categories. L0 is past-directed; centred R³ frames and clip normalization preclude online causality. Tune in development folds; test log loss and calibration on held-out listeners and talkers.

9. Optional neural extension

After behavioral feasibility, separate continuous-speech blocks may test target-versus-masker EEG tracking under acoustically counterbalanced conditions. The rhythm and neural-tracking preprint supplies a relevant precedent,⁵ not proof that improved tracking mediates rehabilitation. EEG remains secondary to recognition and transfer.

Available fNIRS could address slower prefrontal effort-related responses, with short-separation and physiological controls.⁸ It cannot identify a 150-ms attention window. Neither modality is needed for the primary policy trial.

10. Decisions before registration

An independent pilot will establish stimulus quality, profile precision and task burden. Simulation-based sample sizing must cover both the component interactions and the retained between-arm SRT contrast. Fix meaningful-effect bounds, stopping rules, exclusions, multiplicity control and missing-data handling before confirmatory enrollment.

Obtain ethics approval, voice-use permission and licensed assessment materials. The bank generator, feature definitions, model versions and selection policy must be versioned. The present interface is a demonstrator: no participant, generation or MI-extraction results are reported.