Improve Speech Intelligibility Research Articles

Speech perception as afunction of cochlear coverage-comparison in bimodally hearing cochlear implant patients. German version

Hearing success in bimodally hearing patients with acochlear implant (CI) and ahearing aid (HA) exhibits different results: while some benefit from bimodal CI and HA, others do not. The aim of this study was to investigate hearing success in terms of speech perception in bimodally fitted patients in relation to the cochlear coverage (CC) of the CI electrodes. Using the OTOPLAN software (CAScination AG, Bern, Switzerland), CC was retrospectively measured from CT scans of the temporal bone of 39patients, who were then categorized into two groups: CC ≤ 65% (CC500) and CC > 65% (CC600). Monaural speech intelligibility for monosyllables at asound pressure level (SPL) of 65 dB in open field was assessed at various timepoints, preoperatively with HA and postoperatively with CI, and compared between the groups. In addition, speech intelligibility was correlated with CC in the entire cohort before surgery and during follow-up (FU). Overall, no significant differences in speech intelligibility were found between CC500 and CC600 patients at any of the FU timepoints. However, both CC500 and CC600 patients showed asteady improvement in speech intelligibility after implantation. While CC600 patients tended to show an earlier improvement in speech intelligibility, CC500 patients tended to show aslower improvement during the first 3months and asteeper learning curve thereafter. The two patient groups converged during FU, with no significant differences in speech intelligibility. There was no significant relationship between unimodal/unilateral free-field speech intelligibility and CC. However, patients with aCC of 70-75% achieved maximum speech intelligibility. Despite anonsignificant correlation between CC and speech discrimination, patients seem to reach their maximum in unimodal/unilateral speech understanding mainly at 70-75% coverage. However, there is room for further investigation, as CC500 was associated with ashorter cochlear duct length (CDL), and long and very long electrodes were used in both groups.

Speech Intelligibility and Speech Naturalness while Speaking with and without Medical Mask

Introduction: The COVID-19 situation has led to an increase use of medical mask for protection. Facial and respiratory PPE covering the mouth and nose have been reported to diminish Speech Intelligibility. Altering the volume, rate and volume and rate together may contribute to influencing the speech intelligibility and speech naturalness rating. Therefore, identifying the appropriate method to speak with medical mask can help train people to communicate with mask. This study aimed to (1) analyse effects of wearing masks on speech intelligibility, naturalness and comprehensibility, and (2) document an altered style of speaking to improve speech intelligibility, naturalness and comprehensibility while wearing masks. Method: Cross-over study design was used. Spoken utterances from 14 naïve speakers were recorded with and without medical mask, and speech in typical style, increased volume (loudness), reduced rate, and mixed method (combination of increased loudness and reduced rate) with medical mask was recorded from 12 professional speakers. Two expert listeners rated the naïve speakers, and two naïve listeners rated professional speakers using standard rating scales of speech intelligibility and naturalness. Percentage of correct identification of utterance was calculated as a measure of comprehensibility. Wilcoxon signed rank test and Friedman test were used to report significant differences among conditions and parameters analysed. Results: Speech intelligibility, naturalness and comprehensibility was poorer while speaking with medical mask than without them. Speaking in mixed method (combination of increased loudness and reduced rate), while using medical masks was the most appropriate style of speech in order to be intelligible, natural and comprehensible, followed by increased loudness style, typical style and reduced rate style. Conclusion: People should be alerted that wearing mask leads to reduction in speech intelligibility, comprehensibility and naturalness. These can be improved by modifying the style of speaking.

Avocodo: Generative Adversarial Network for Artifact-Free Vocoder

Neural vocoders based on the generative adversarial neural network (GAN) have been widely used due to their fast inference speed and lightweight networks while generating high-quality speech waveforms. Since the perceptually important speech components are primarily concentrated in the low-frequency bands, most GAN-based vocoders perform multi-scale analysis that evaluates downsampled speech waveforms. This multi-scale analysis helps the generator improve speech intelligibility. However, in preliminary experiments, we discovered that the multi-scale analysis which focuses on the low-frequency bands causes unintended artifacts, e.g., aliasing and imaging artifacts, which degrade the synthesized speech waveform quality. Therefore, in this paper, we investigate the relationship between these artifacts and GAN-based vocoders and propose a GAN-based vocoder, called Avocodo, that allows the synthesis of high-fidelity speech with reduced artifacts. We introduce two kinds of discriminators to evaluate speech waveforms in various perspectives: a collaborative multi-band discriminator and a sub-band discriminator. We also utilize a pseudo quadrature mirror filter bank to obtain downsampled multi-band speech waveforms while avoiding aliasing. According to experimental results, Avocodo outperforms baseline GAN-based vocoders, both objectively and subjectively, while reproducing speech with fewer artifacts.

Signal Transparency of Remote Microphone Technology in Pediatric Bone Conduction Device Users

Introduction: Competing noise in the environment negatively affects speech intelligibility, particularly when listening at a distance. This is especially true for children with hearing loss in classroom environments where the signal-to-noise ratio is often poor. Remote microphone technology has been shown to be highly beneficial at improving the signal-to-noise ratio in hearing device users. Children with bone conduction devices, however, often must rely on indirect transmission of the acoustic signal for commonly used classroom-based remote microphone listening (e.g., digital adaptive microphone) which may negatively affect speech intelligibility. There are no studies on the effectiveness of using a relay method of signal delivery using remote microphone technology to improve speech intelligibility in adverse listening environments in bone conduction device users. Methods: Nine children with irresolvable conductive hearing loss and 12 adult controls with normal hearing were included for study. Controls were bilaterally plugged to simulate conductive hearing loss. All testing was conducted using the Cochlear™ Baha® 5 standard processor coupled with either the Cochlear™ Mini Microphone 2+ digital remote microphone or the Phonak Roger™ adaptive digital remote microphone. Speech intelligibility in noise was evaluated in the (1) bone conduction device processor alone, (2) bone conduction device + personal remote microphone, and (3) bone conduction device + personal remote microphone + adaptive digital remote microphone listening conditions at −10 dB, 0 dB, and +5 dB signal-to-noise ratios. Results and Conclusions: Speech intelligibility in noise improved significantly in the bone conduction device + personal remote microphone condition over the bone conduction device alone, demonstrating significant benefit for listening at poor signal-to-noise ratios in children with conductive hearing loss using bone conduction devices with personal remote microphone use. Experimental findings demonstrate poor signal transparency when using the relay method. Coupling of the adaptive digital remote microphone technology to the personal remote microphone negatively affects signal transparency, and no hearing in noise improvements are observed. Significant gains in speech intelligibility are consistently observed for direct streaming methods and are confirmed in adult controls. Behavioral findings are supported by objective verification of the signal transparency between the remote microphone and the bone conduction device.

Numerical simulation of on-road wind conditions for interior wind noise of passenger vehicles

Traditionally vehicles are designed for wind noise under ideal steady wind conditions. But, passenger comfort is affected by high modulation of cabin noise while cruising in traffic due to variations of instantaneous wind speed and direction from driving through large-scale turbulence. In consequence, designing a vehicle for the best performance in a low-turbulence wind tunnel may lead to issues during on-road conditions. To predict the interior noise corresponding to on-road turbulence, a simulation approach is proposed combining an upstream turbulence flow simulation with an SEA vehicle model. This work is an extension of existing well validated procedures for steady wind conditions. Time-segmented transient loads on panels and steady-state structural acoustics transfer functions are combined, producing interior noise results for a series of overlapping time segments. This interior noise prediction, as a function of time, captures the modulation of wind noise results, which are then visualized using a spectrogram, plotting sound pressure level indicated by color as a function of time and frequency bands. Audio signals are also created from the interior noise results, allowing detailed signal analysis from this simulation. This capability enables designers to reduce wind noise modulation in time caused by on-road turbulence, improving speech intelligibility and passenger comfort.

Improving Speech Intelligibility from PA Systems in Transportation Terminals

ACRP 175 was published in 2018 and provides basic guidelines for design, installation, and operation of public address systems in airport terminals. There are approximately 400 airports in the US, and with over 550 underground stations and 2,600 above ground stations in the US , many of these guidelines developed for airports can also be easily adapted for these transit stations. This paper provides an overview of ACRP 175 guidance with a focus on key factors for successful design and commissioning. As part of the research acoustic field studies were performed at six airports and a total of 45 different spaces including ticketing, concourse, gate lounges, food courts, arrival/departure halls, baggage claim, and curbside locations. We describe experience using ACRP 175 on current design-build projects and discuss other relevant industry standards.

Improving communications in PPE: a solution for ‘landline’ telephone communication

BackgroundEmergency care staff wearing elastomeric respiratory personal protective equipment (PPE) report difficulties in communicating by telephone. We developed and tested an affordable technological solution aimed at improving telephone call intelligibility...

Tonotopic fitting of the sound processor cochlear implant in normal cochlea anatomy

Introduction. Cochlear implantation is very effective due to the known tonotopic organization of the cochlea. These data are used in the distribution of signals along the channels of the cochlear implant electrode. The more accurate and natural the stimulation of the auditory nerve fibers, the better the perception of speech, sounds and speech intelligibility.The aim of the study was to compare the clinical and anatomical settings of the cochlear implant processor.Materials and methods. The study included 63 patients aged 2 to 60 years using cochlear implantation system, implants with a standard long electrode array (31.5 mm). All participants underwent computed tomography (CT) of the temporal bones with a step of 0.6 mm or less. CT data were processed using Otoplan and Sliser 3D software. The geometric dimensions of the cochlea, the length of the cochlear canal, the angular position and tonotopic frequency of each electrode were calculated, and a 3D reconstruction of the cochlea and electrode was built.Results. The correction of the frequency filters of the electrodes and the anatomical adjustment of the processor were performed according to the data obtained in the Otoplan program. A comparison was made between the clinical and anatomical setting. After the redistribution of frequency filters, patients noted a more natural sound, improved speech intelligibility. Thus, with anatomical adjustment, an exact correspondence is achieved between the central frequency of each electrode channel and the tonotopic frequency of the cochlear zone, individually determined by computed tomography.Conclusions. A new tool has appeared that allows you to significantly improve the quality and provide an individual approach to setting up processors after cochlear implantation.

Matching the pitch perception of the cochlear implanted ear with the contralateral ear in patients with single-sided deafness: a novel approach.

Single-sided deaf patients following cochlear implantation often compare the sound quality of their implanted ear with normal hearing. The interaural differences can result in dissatisfaction with speech comprehension and reduced time of usage of the speech processor; hence, prolonging auditory adaptation time. The proposed calibration method presented in this study demonstrates how the frequency distribution of the cochlear implant can be set to adequately approximate the pitch perception of the contralateral normal hearing ear towards improving speech intelligibility in a noisy environment. In 12 postlingual single-sided deaf patients, subjective interaural pitch-matching was carried out to determine new central frequencies for the reallocation of the frequency bands of their speech processor (CP910, CP950 or CP1000, Cochlear, Australia). The patients were asked to compare the pitch of the tones presented to their normal hearing ear to the pitch of individual channels of their cochlear implant (CI522 or CI622, Cochlear, Australia). A third-degree polynomial curve was fit to the acquired matching frequencies to create the new frequency allocation table. Audiological measurements (free-field aided thresholds, speech reception thresholds, and monosyllabic word recognition score) in noise, together with a Speech, Spatial and Qualities of Hearing Scale (SSQ12) questionnaire (short version of the original SSQ) results were evaluated prior to the pitch-matching procedure, and again, 2 weeks later. The free-field aided thresholds of the patients showed no greater shift than ± 5dB following the procedure; however, their monosyllabic word recognition score in noise improved significantly (mean - 9.58%, SD 4.98%, matched pairs t test comparison: p < 0.001). The results of the SSQ12 questionnaire also showed significant improvement in speech intelligibility, sound localization, and sound quality (mean 0.96 points, SD 0.45 points, matched pairs t test comparison: p < 0.001). Matching the pitch perception of the implanted cochlea with the sensation of the normal hearing contralateral ear, resulted in significant changes in the quality of hearing in patients with single-sided deafness. It is plausible the procedure can usher positive results in bimodal patients or following sequential bilateral cochlear implantation.

THE ASSESSMENT OF ACOUSTICAL PERFORMANCE OF MOSQUES IN SHAH ALAM

A mosque is a place where Muslims conduct activities such as praying and listening to talks. Therefore, it is important for a mosque to have great acoustics quality as it influences the worshipper’s experience when praying or listening to Imam’s preaches. The acoustical evaluation in the mosque is rarely conducted and, in most cases, the development of good acoustic quality in the mosque is never considered at an early stage of design. In the present investigation, two large-volume mosques in Shah Alam were investigated: Section 7 Mosque (MS7) and Sultan Salahuddin Abdul Aziz Mosque (MSA). The objective of this study is to evaluate the acoustical characteristics of the mosque and its quality level. In addition, the speech intelligibility performance of each mosque was also determined using Speech Transmission Index (STI) and Rapid Transmission Index (RASTI). The experimental work found that the reverberation time of MSA has attained recommended values of reverberation time for that given size and volume while MS7 has slightly higher. The speech level attainment of both mosques is sufficient for worshippers to hear comfortably, but the background noise is high. Nevertheless, the signal-to-noise ratio (S/N ratio) indicates that both mosques have been able to obtain more than 15 dB(A) for improved speech intelligibility. The STI and RASTI prediction shows that MS7 has attained ‘fair’ and ‘bad’ ratings. In contrast, ‘poor’ and ‘bad’ ratings are achieved in the MSA.

Binaural pre-processing for contralateral sound field attenuation can improve speech-in-noise intelligibility for bilateral hearing-aid users

We have recently proposed a binaural sound pre-processing method to attenuate sounds contralateral to each ear and shown that it can improve speech intelligibility for normal-hearing (NH) people in simulated “cocktail party” listening situations (Lopez-Poveda et al., 2022, Hear Res 418:108,469). The aim here was to evaluate if this benefit remains for hearing-impaired listeners when the method is combined with two independently functioning hearing aids, one per ear. Twelve volunteers participated in the experiments; five of them had bilateral sensorineural hearing loss and seven were NH listeners with simulated bilateral conductive hearing loss. Speech reception thresholds (SRTs) for sentences in competition with a source of steady, speech-shaped noise were measured in unilateral and bilateral listening, and for (target, masker) azimuthal angles of (0°, 0°), (270°, 45°), and (270°, 90°). Stimuli were processed through a pair of software-based multichannel, fast-acting, wide dynamic range compressors, with and without binaural pre-processing. For spatially collocated target and masker sources at 0° azimuth, the pre-processing did not affect SRTs. For spatially separated target and masker sources, the pre-processing improved SRTs when listening bilaterally (improvements up to 10.7 dB) or unilaterally with the acoustically better ear (improvements up to 13.9 dB), while it worsened SRTs when listening unilaterally with the acoustically worse ear (decrements of up to 17.0 dB). Results show that binaural pre-processing for contralateral sound attenuation can improve speech-in-noise intelligibility in laboratory tests also for bilateral hearing-aid users.

The effect of explicit and implicit voice training on speech-on-speech intelligibility and listening effort

Listening to familiar voices might improve intelligibility of target speech in multiple-talker situations. In addition to personally familiar voices, voice training—through implicit exposure or explicit learning of a previously unheard voice—can improve speech intelligibility. However, there is no consensus on which method is more effective. We investigated the effect of explicit and implicit voice training on speech-on-speech perception and listening effort (pupillometry), among normal hearing listeners. There was no significant difference in speech intelligibility performance between trained and untrained voices, with either voice training method. However, an interaction between voice training method and target-to-masker ratio (TMR) on accuracy scores showed that the benefit of increased TMR on accuracy scores was significantly larger when voice training was implicit compared to explicit. Pupillometry results showed that explicit voice training significantly reduced listening effort for most disadvantageous TMR (−6dB), but not for more advantageous TMRs (0 and + 6 dB), and not for implicitly trained voices. Our results may imply that explicit voice training is more effective than implicit voice training in reducing listening effort. Still, implicit voice training, perhaps via procedural learning, might be more beneficial on the performance level when TMR becomes more advantageous.

The impact of hearing aid user's own voice on device signal processing

Hearing aid processing is designed to improve audibility for sounds of interest, often by targeting external speech signals. During natural conversation, however, the hearing aid user is also the source of speech, potentially interacting with hearing aid function and leading to suboptimal processing. In this study, we investigated how the presence of own voice impacts the deployment of specific features designed to enhance audibility and intelligibility of conversational partners. We recorded real-time hearing aid feature engagement (directional microphones, noise reduction, and speech enhancement) during simulated conversations in quiet and in background noise. Simulations were performed using an acoustic manakin (GRAS 45BC-12) capable of producing speech via internal speaker, which was situated in the center of a 24-speaker spatial array. The results demonstrate how the presence of a hearing aid user’s own voice disrupts the intended use of these adaptive features, and that there is a need for optimizing hearing enhancement devices to account for own voice in dynamic scenarios. Future studies will determine the impact of a hearing aid user’s own voice on the device’s ability to improve speech intelligibility and user satisfaction.

Impact of clear face masks on audiovisual speech intelligibility and subjective listening effort with normal hearing young adults

During the COVID-19 pandemic, facemask use made people suddenly aware of the importance of both visual information and acoustic clarity in speech perception. Masks with transparent panels can provide listeners with visual speech information, but such masks tend to introduce more acoustic distortion than other opaque masks. Thus, their value for improving speech recognition (at least for listeners with normal hearing) is not clear. In the current study we investigated the ability of three different clear face masks to improve speech intelligibility relative to an opaque surgical mask. Normal hearing young adult participants (N = 180) were presented with audiovisual recordings of 150 sentences from a female speaker in 5 different mask conditions (no mask, surgical mask, and three different clear masks) and at three different signal-to-noise ratios (no noise, –5 dB SNR, and −9 dB SNR). After each condition, participants also completed the NASA-TLX survey for subjective effort. Participants also completed an independent lipreading task. Preliminary data suggest decreased subjective effort and improved performance of the clear masks relative to the opaque mask at high SNR. This implies a potential for visual information to make up for decreased acoustic clarity, making clear masks a better choice in difficult listening conditions.

Characteristics of Alveolo-palatal Affricates Produced by Mandarin-speaking Children with Repaired Cleft Palate.

Objectives: In this study, examined the acoustic properties of affricates /t/ and /th/ in Mandarin Chinese, and analyzed the differences of the acoustic characteristics of these affricates produced by children with repaired cleft palate and normally developing children. We also explored the relationship between the affricates and high-front vowel /i/. Methods: We analyzed 16 monosyllabic words with alveolo-palatal affricates as the initial consonants produced by children with repaired cleft palate (N=13, Mean=5.9 years) and normally developing children (N=6, Mean age=5.3 years). We used several acoustic parameters to investigate the characteristics of these affricates, such as the center of gravity, VOT and the formants of vowels. Results: Compared with normally developing children, children with cleft palate exhibited a lower center of gravity for the 2 affricates /t/ and /th/. Data from the control group showed that the affricate /th/ had a significantly greater center of gravity than that of /t/. The accuracy of /t , th/ produced by speakers of cleft palate was significantly correlated with that of /i/ (r=0.63). High-front vowel /i/ is a significant index in diagnosing speech intelligibility which is more valuable than /a/ and /u/. There was a significant difference in F2 of vowel /i/ between children with cleft palate without speech therapy (CS1) and after speech therapy (CS2). After speech intervention, the accuracy of affricates produced by children with cleft palate was improved, the acoustic properties "stop + noise segments" appeared. Conclusion: Children with cleft palate can be distinguished better from children with normal development by 2 significant acoustic characteristics: center of gravity and VOT. As alveolo-palatal affricates /t , th/ and high-front vowel /i/ have a similar place of articulation, front-tongue-blade, their production accuracy can be improved mutually. The analysis showed that the articulation of Chinese /i/ has a higher frontal lingual position and less variability, which is more conducive to articulation training and improves the effect of cleft palate training. These findings provide a potential relationship on affricates /t, th/ and vowel /i/. Children with cleft palate have difficulty pronouncing the /t, t h/ and /i/. It is better to start with a vowel /i/, resulting in improvement in overall speech intelligibility.

CST: Complex Sparse Transformer for Low-SNR Speech Enhancement.

Speech enhancement tasks for audio with a low SNR are challenging. Existing speech enhancement methods are mainly designed for high SNR audio, and they usually use RNNs to model audio sequence features, which causes the model to be unable to learn long-distance dependencies, thus limiting its performance in low-SNR speech enhancement tasks. We design a complex transformer module with sparse attention to overcome this problem. Different from the traditional transformer model, this model is extended to effectively model complex domain sequences, using the sparse attention mask balance model's attention to long-distance and nearby relations, introducing the pre-layer positional embedding module to enhance the model's perception of position information, adding the channel attention module to enable the model to dynamically adjust the weight distribution between channels according to the input audio. The experimental results show that, in the low-SNR speech enhancement tests, our models have noticeable performance improvements in speech quality and intelligibility, respectively.

FDTD-Based Simulation and Analysis of Noise Reduction Effects in Hospital Wards

The impact of domestic hospital acoustic environment on the health of patients and healthcare workers is attracting more and more attention. This paper investigated and analyzed the noise situation of typical double wards in hospital, and designed noise reduction schemes according to its noise sources; based on finite difference time domain method (FDTD), sound fields in wards were simulated before and after noise reduction treatments, and noise reduction results were compared and analyzed. Results show that FDTD can be used to effectively model sound field of wards with complex boundary conditions; meanwhile, the noise mitigation processes have significant effects on improvement of reverberation time and speech intelligibility in wards.

Predicting Speech Intelligibility for People with a Hearing Loss: The Clarity Challenges

Objective speech intelligibility metrics are used to reduce the need for time consuming listening tests. They are used in the design of audio systems; room acoustics and signal processing algorithms. Most published speech intelligibility metrics have been developed using young adults with so-called 'normal hearing', and therefore do not work well for those with different hearing characteristics. One of the most common causes of aural diversity is sensorineural hearing loss. While partially restoring perception through hearing aids is possible, results are mixed. This has led to the Clarity Project, which is running an open series of Enhancement Challenges to improve the processing of speech-in-noise for hearing aids. To enable this, objective metrics of speech intelligibility are needed, which work from signals produced by hearing aids for diverse listeners. For this reason, Clarity is also running Prediction Challenges to improve speech intelligibility models. Competitors are given a set of audio signals produced by hearing aid algorithms, and challenged to predict how many words a listener with a particular hearing characteristic will achieve. Drawing on the learning from the challenge, we will outline what has been learnt about improving intelligibility metrics for those with a hearing impairment.

Hearing rehabilitation and microbial shift after middle ear surgery with Vibrant Soundbridge in patients with chronic otitis media

IntroductionPatients with otitis media (OM) encounter significant functional hearing impairment with conductive, or a combined hearing loss and long-term sequelae involving impaired speech/language development in children, reduced academic achievement and irreversible disorders of middle and inner ear requiring a long time therapy and/or multiple surgeries. In its persistent chronic form, Otitis media (COM) can often only be treated by undergoing ear surgery for hearing restoration. The persistent inflammatory reaction plays a major role, often caused by multi-resistant pathogens in the ear. Herein, we present outcomes of patients implanted with currently the only FDA approved active Middle Ear Implant Vibrant Soundbridge (VSB), suffering from persistent COM.MethodsThe study enrolled 42 patients, treated by performing middle ear (ME) surgery to different extents and implanted with the VSB to various structures in the ME. Included were 17 children and 25 adults that had recurrent and/or persisting OM and significant hearing loss. Preoperative and postoperative patients' audiometric data were evaluated and the benefit with VSB assessed using the Glasgow Benefit Inventory for adults and pediatric cohorts. The microbial spectrum of pathogens was assessed before and after surgery, exploring the colonization of the otopathogens, as well as the intestinal microbiome from individually burdened patients.ResultsThe mean functional gain is 29.7 dB HL (range from 10 to 56.2 dB HL) with a significant improvement in speech intelligibility in quiet. Following VSB implantation, no significant differences in coupling were observed at low complication rates. Postoperatively patients showed significantly increased benefit with VSB compared to the untreated situation, including less otorrhea, pain, medical visits, and medication intake, with no recurrent OM and significant bacterial shift in otopathogens. The analysis of the intestinal microbiome displayed a high abundance of bacterial strains that might be linked to chronic and persistent inflammation.ConclusionsFunctional ear surgery including rehabilitation with a VSB in patients suffering from COM present to be safe and effective. The successful acceptance accompanied by the improved audiological performance resulted in significant benefit with VSB, with a shift in the ear pathogens and altered microbiome and thus is a great opportunity to be treated.

Motor Speech Interventions for Children With Cerebral Palsy: A Systematic Review.

Dysarthria is common among children with cerebral palsy (CP) and results in poor speech intelligibility and subsequently low communicative participation. Neuroplasticity evidence suggests that adherence to principles of motor learning (PML) improves motor speech intervention outcomes. Motor speech interventions aim to improve speech intelligibility and ultimately facilitate participation, but the effectiveness of these interventions and their inclusion of PML are not clear. Aims are as follows: (a) evaluate the effectiveness motor speech interventions in improving speech intelligibility; (b) summarize the aims, outcome measures, and outcomes relating to the International Classification of Functioning and Disability Child and Youth; and (c) summarize the principles of motor learning used in the intervention. Eight databases were searched, complemented by a hand search. Studies of any level of evidence were included if they used a motor speech intervention and measured speech in children with CP aged 0-18 years. Studies before 2000 or not in English were excluded. The review was conducted and reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Study quality was rated using the Single Case Experimental Design Scale and Physiotherapy Evidence Database-P rating scales. The strength of evidence was evaluated with Grading of Recommendations, Assessment, Development and Evaluation. Of 1,036 initial articles, 21 were included. Eight interventions were identified including 131 participants aged 3-18 years. All studies aimed to improve speech intelligibility or articulation and reported improvement at sound, word, or sentence level. One study reported improvements in communicative participation. The strength of evidence ranged from very low to moderate for one intervention. Adherence to PML was inconsistent. The quality of evidence is very low to moderate. More research on motor speech interventions that adhere to PML is required. https://doi.org/10.23641/asha.21817959.