Speech Transcription Research Articles

Cognitive decline assessment using semantic linguistic content and transformer deep learning architecture.

Dementia is a cognitive decline that leads to the progressive deterioration of an individual's ability to perform daily activities independently. As a result, a considerable amount of time and resources are spent on caretaking. Early detection of dementia can significantly reduce the effort and resources needed for caretaking. This research proposes an approach for assessing cognitive decline by analysing speech data, specifically focusing on speech relevance as a crucial indicator for memory recall. This is a cross-sectional, online, self-administered. The proposed method used deep learning architecture based on transformers, with BERT (Bidirectional Encoder Representations from Transformers) and Sentence-Transformer to derive encoded representations of speech transcripts. These representations provide contextually descriptive information that is used to analyse the relevance of sentences in their respective contexts. The encoded information is then compared using cosine similarity metrics to measure the relevance of uttered sequences of sentences. The study uses the Pitt Corpus Dementia dataset for experimentation, which consists of speech data from individuals with and without dementia. The accuracy of the proposed multi-QA-MPNet (Multi-Query Maximum Inner Product Search Pretraining) model is compared with other pretrained transformer models of Sentence-Transformer. The results show that the proposed approach outperforms the other models in capturing context level information, particularly semantic memory. Additionally, the study explores the suitability of different similarity measures to evaluate the relevance of uttered sequences of sentences. The experimentation reveals that cosine similarity is the most appropriate measure for this task. This finding has significant implications for the early warning signs of dementia, as it suggests that cosine similarity metrics can effectively capture the semantic relevance of spoken language. The persistent cognitive decline over time acts as one of the indicators for prevalence of dementia. Additionally early dementia could be recognised by analysis on other modalities like speech and brain images. What is already known on this subject It is already known that speech- and language-based detection methods can be useful for dementia diagnosis, as language difficulties are often early signs of the disease. Additionally, deep learning algorithms have shown promise in detecting and diagnosing dementia through analysing large datasets, particularly in speech- and language-based detection methods. However, further research is needed to validate the performance of these algorithms on larger and more diverse datasets and to address potential biases and limitations. What this paper adds to existing knowledge This study presents a unique and effective approach for cognitive decline assessment through analysing speech data. The study provides valuable insights into the importance of context and semantic memory in accurately detecting the potential in dementia and demonstrates the applicability of deep learning models for this purpose. The findings of this study have important clinical implications and can inform future research and development in the field of dementia detection and care. What are the potential or actual clinical implications of this work? The proposed approach for cognitive decline assessment using speech data and deep learning models has significant clinical implications. It has the potential to improve the accuracy and efficiency of dementia diagnosis, leading to earlier detection and more effective treatments, which can improve patient outcomes and quality of life.

Children With Developmental Language Disorder Show Deficits in the Production of Musical Rhythmic Groupings.

Children with developmental language disorder (DLD) show evidence of domain-general deficits in sequentially patterned motor skills. This study focuses on the production of rhythmically grouped sequences drawn from a music task, with the hypothesis that children with DLD will show a sequential pattern learning deficit that crosses language and action domains. Fifty-seven 4- to 5-year-old children (36 with DLD) drummed and clapped a developmentally appropriate musical rhythmic sequence 24 times (clapped 12 times, drummed 12 times). The accuracy of rhythmic events (markings of claps, drums, and pauses in a target sequence) was assessed through a modification of classic speech and language transcription procedures. The variability and prosodic structure of the rhythmic groupings were also measured. Children with DLD produced less accurate and more variable rhythmic groupings compared to their typically developing (TD) peers. While the final-position grouping of the sequence was especially vulnerable for all children, those with DLD included more co-occurring errors in initial and final groupings of the same rhythmic sequence. Both TD children and children with DLD were less accurate in the clapping than the drumming task. Neither rhythmic drumming nor clapping accuracy correlated with motor skill in either group of children. This study provides novel evidence of a manual rhythmic grouping deficit in DLD, one that is motivated by language-not motor or speech-factors. Cognitive abilities necessary to organize rhythmic events into higher order groupings are impaired across music and language in children with DLD. Rhythmic organization and sequencing may serve an important role in diagnosis and intervention in this population. https://doi.org/10.23641/asha.24158745.

18 Automatically calculated lexical and sentential context features of connected speech predict cognitive impairment

Objective:Early detection is critical to the effective management of Alzheimer’s Disease (AD) and other forms of dementia. Existing screening assessments are often costly, require substantial expertise to administer, and may be insensitive to mild changes in cognition. A promising alternative is automatically measuring features of connected speech (c.f., Ostrand & Gunstad, 2021, Journal Geriatric Psych & Neurol) to predict impairment. Here, we built on prior work examining how well speech features predicted cognitive impairment. Unique to the current work, we attempted to capture more holistic effects of cognitive impairment by examining the relevance of linguistic features that measure sentential or discourse context properties of speech, including the context in which filler words (e.g., um) occur, and the predictability of individual words within their sentence context, computed from a large computational language model (GPT-2).Participants and Methods:Participants completed the Cookie Theft picture description task, with data available in the DementiaBank corpus (Becker et al., 1994, Arch Neurol). Descriptions that contained at least 50 words (N = 214) were submitted to an automatic feature calculation pipeline written in Python to calculate various part-of-speech counts, lexical diversity metrics, and mean lexical frequency, as well as multiple metrics related to lexical surprisal (i.e., how surprising a word is given its context).Surprisal of individual words was computed using the pre-trained GPT-2 transformer language model (Radford et al., 2019, Comput. Sci.) by computing word probability given the previous 12 words. Multiple linear regression was performed using 17 linguistic features jointly as predictors, and Mini-Mental State Examination (MMSE) score as the outcome variable. Simple regressions were calculated between each feature and MMSE scores to examine the predictability of each linguistic feature on cognitive decline.Results:A multiple linear regression model containing all linguistic features plus demographic information (age, sex, education) significantly predicted MMSE scores (Adjusted R2 = 0.41, F20, 193 = 8.37, p < .001), and explained significantly more variance in MMSE scores than did demographic variables alone (F17193 = 6.85, p < .001). Individual predictors that were significantly correlated with MMSE score included: how unexpected an individual’s word choices were, given the preceding context (median surprisal: r = -0.33, p < 0.001; interquartile range: r = 0.18, p < 0.02), mean lexical frequency (r = -0.50, p < .001), and usage of definite articles (r = 0.31, p < 0.001), nouns (r = 0.26, p < .001), and empty words (r = -0.25, p < 0.001).Conclusions:Participants with lower MMSE scores, indicating greater impairment, used more frequent, yet more surprising, words, and produced more empty words and fewer definite articles and nouns. These results suggest that measures of semantic specificity and coherence of speech could be meaningful predictors of cognitive decline, and can be computed automatically from speech transcriptions. The results also provide novel evidence that computational approaches to estimating lexical predictability may have value in predicting the degree of decline, motivating future work in other speech elicitation tasks and differing clinical groups.

20 Global and Local Semantic Coherence of Spontaneous Speech in Persons with Alzheimer's Disease and Healthy Controls

Objective:Growing evidence demonstrates that subtle changes in spontaneous speech can be used to distinguish older adults with and without cognitive impairment, including those with Alzheimer's disease (AD). Recent work suggests that quantification of the meaningful connectedness of speech - termed semantic coherence - may be sensitive to cognitive dysfunction. The current study compared global coherence (GC; the degree to which individual utterances relate to the overall topic being discussed) and local coherence (LC; the degree to which adjoining utterances relate meaningfully to each other) in persons with AD and healthy controls.Participants and Methods:Speech transcripts from 81 individuals with probable AD (Mage = 72.7 years, SD = 8.8, 70.3% female) and 61 healthy controls (HC) (Mage = 63.9 years, SD = 8.5, 62.2% female) from Dementia Bank were analyzed. All participants completed the Cookie Theft and MMSE as part of that larger project. Machine learning analyses of GC and LC were conducted and models evaluated classification accuracy (i.e., AD vs HC) as well as ROC-AUC. Relationships between coherence indices and MMSE performance were also quantified.Results:Though no significant group differences emerged in LC (Estimate = 0.012, p = 0.32), persons with AD differed from healthy controls in GC (Estimate = 0.03, p = 0.006) and produced less semantically coherent speech. GC indices predicted AD diagnoses with 65% accuracy. Interestingly, coherence indices showed only modest correlation with MMSE scores (r = .19).Conclusions:GC metrics of spontaneous speech differentiated between persons with AD and controls, but did not strongly correlate with MMSE performance. Such findings support the notion that many aspects of language are impacted in persons with AD. In addition to replication, future work should evaluate whether GC is also disrupted in persons with pre-clinical AD and its potential to assist with early detection.

On the possibilities of introducing transcription into the practice of interrogation at the stage of preliminary investigation of crimes

The subject of the study is the modern forensic investigative practice of recording the testimony of interrogated persons at the preliminary investigation, which indicates a tendency to increase the volume of interrogation protocols being compiled. The object of the study is the regularities of the procedure for recording the testimony of the interrogated persons. It is noted that the preparation of the interrogation protocol is the most time-consuming final stage of this investigative action. Analysis of scientific publications in recent years shows that forensic scientists focus on the tactics and psychology of interrogation. As part of the forensic tactics of interrogation, the issues of the specifics of the use of additional means of fixation in the form of audio-video recordings and requirements for the content of the interrogation protocol are also considered. The existing procedure for recording the testimony of interrogated persons in the typewriter mode remains acute. The novelty of the research lies in the author's proposal to introduce modern computer technologies and artificial intelligence into the practice of drawing up interrogation protocols that allow converting oral speech into text. We are talking about automatic speech recognition and the conversion of a speech signal into digital information in the form of text (transcription of oral speech). Currently, transcription is widely used in various spheres of human activity. It is noted that it is permissible to use transcription technology from the standpoint of criminal procedural regulation of interrogation and confrontation.The use of transcription means will allow the investigator to focus on communicating with the interrogated person, achieve psychological contact, track testimony, use tactical techniques, etc. Modern software allows the introduction of the specified technology of recording the testimony of interrogated persons into the practice of interrogation.

Mood and Modality in Interpersonal Meaning of Joseph Robinette Biden Victory Speech

Speech involves the articulation of one's ideas and emotions. Politicians must possess the skill to effectively utilize language. Prior research underscores the significance of studying the political language, as it empowers individuals to grasp how those aspiring to attain, wield, and retain authority employ language. The purpose of this study is to analyze how mood and modality were used in the victory speech. This research applies qualitative design. The transcription of Joe Biden's victory speech is the subject of this investigation. Mr. Biden convey commitments and responsibility connected with the program that he will run; however, we don't know whether it will be run or no. This research applied qualitative research. The analysis is using data collection, data reduction, data display, and conclusion. According to the findings of this study, there are three distinct categories: value, modality, and mood. Declarative mood is the most common type; 193 clauses, or about 94%, are in the imperative mood, followed by seven clauses, or 3,4%, and three clauses, or about 1,5 percent, in the interrogative mood. The dominant modality value is median, there are eighteen provisions or around 62,1%, then, at that point, trailed by eight clauses or around 27,6% as low worth and the latter is three conditions or 10,3% for high worth. The dominant type of modality is obligation. Eleven clauses, or approximately 37.9 percent, are classified as obligation, ten clauses, or approximately 34.5%, as probability, and eight clauses, or approximately 26.7 percent, as inclination.

Approach to automatic recognition of emotions in speech transcriptions

Approach to automatic recognition of emotions in speech transcriptions

Kim Namjoon Speech at the UN Assembly: A Discourse Analysis

This research aims at revealing values, belief and assumptions are found in Kim Namjoon’s speech text on campaigning “Love Yourself” and to know types of figurative language used in the speech. This research used descriptive qualitative method. The instrument used in this research was the speech text transcript and speech video that retrieved from YouTube. Data collection techniques used in this research was library research techniques, thus seeing and recording as the main approach to collect the data. The qualitative data analysis was employed after following stages; analyzing data reduction, data display and verification.

Radio2Text

Millimeter wave (mmWave) based speech recognition provides more possibility for audio-related applications, such as conference speech transcription and eavesdropping. However, considering the practicality in real scenarios, latency and recognizable vocabulary size are two critical factors that cannot be overlooked. In this paper, we propose Radio2Text, the first mmWave-based system for streaming automatic speech recognition (ASR) with a vocabulary size exceeding 13,000 words. Radio2Text is based on a tailored streaming Transformer that is capable of effectively learning representations of speech-related features, paving the way for streaming ASR with a large vocabulary. To alleviate the deficiency of streaming networks unable to access entire future inputs, we propose the Guidance Initialization that facilitates the transfer of feature knowledge related to the global context from the non-streaming Transformer to the tailored streaming Transformer through weight inheritance. Further, we propose a cross-modal structure based on knowledge distillation (KD), named cross-modal KD, to mitigate the negative effect of low quality mmWave signals on recognition performance. In the cross-modal KD, the audio streaming Transformer provides feature and response guidance that inherit fruitful and accurate speech information to supervise the training of the tailored radio streaming Transformer. The experimental results show that our Radio2Text can achieve a character error rate of 5.7% and a word error rate of 9.4% for the recognition of a vocabulary consisting of over 13,000 words.

Privacy against Real-Time Speech Emotion Detection via Acoustic Adversarial Evasion of Machine Learning

Smart speaker voice assistants (VAs) such as Amazon Echo and Google Home have been widely adopted due to their seamless integration with smart home devices and the Internet of Things (IoT) technologies. These VA services raise privacy concerns, especially due to their access to our speech. This work considers one such use case: the unaccountable and unauthorized surveillance of a user's emotion via speech emotion recognition (SER). This paper presents DARE-GP, a solution that creates additive noise to mask users' emotional information while preserving the transcription-relevant portions of their speech. DARE-GP does this by using a constrained genetic programming approach to learn the spectral frequency traits that depict target users' emotional content, and then generating a universal adversarial audio perturbation that provides this privacy protection. Unlike existing works, DARE-GP provides: a) real-time protection of previously unheard utterances, b) against previously unseen black-box SER classifiers, c) while protecting speech transcription, and d) does so in a realistic, acoustic environment. Further, this evasion is robust against defenses employed by a knowledgeable adversary. The evaluations in this work culminate with acoustic evaluations against two off-the-shelf commercial smart speakers using a small-form-factor (raspberry pi) integrated with a wake-word system to evaluate the efficacy of its real-world, real-time deployment.

Word segmentation from transcriptions of child-directed speech using lexical and sub-lexical cues.

We compare two frameworks for the segmentation of words in child-directed speech, PHOCUS and MULTICUE. PHOCUS is driven by lexical recognition, whereas MULTICUE combines sub-lexical properties to make boundary decisions, representing differing views of speech processing. We replicate these frameworks, perform novel benchmarking and confirm that both achieve competitive results. We develop a new framework for segmentation, the DYnamic Programming MULTIple-cue framework (DYMULTI), which combines the strengths of PHOCUS and MULTICUE by considering both sub-lexical and lexical cues when making boundary decisions. DYMULTI achieves state-of-the-art results and outperforms PHOCUS and MULTICUE on 15 of 26 languages in a cross-lingual experiment. As a model built on psycholinguistic principles, this validates DYMULTI as a robust model for speech segmentation and a contribution to the understanding of language acquisition.

Assessing coherence through linguistic connectives: Analysis of speech in patients with schizophrenia-spectrum disorders.

Incoherent speech is a core diagnostic symptom of schizophrenia-spectrum disorders (SSD) that can be studied using semantic space models. Since linguistic connectives signal relations between words, they and their surrounding words might represent linguistic loci to detect unusual coherence in speech. Therefore, we investigated whether connectives' measures are useful to assess incoherent speech in SSD. Connectives and their surrounding words were extracted from transcripts of spontaneous speech of 50 SSD-patients and 50 control participants. Using word2vec, two different cosine similarities were calculated: those of connectives and their surrounding words (connectives-related similarity), and those of free-of-connectives words-chunks (non-connectives similarity). Differences between groups in proportion of five types of connectives were assessed using generalized logistic models, and connectives-related similarity was analyzed through non-parametric multivariate analysis of variance. These features were evaluated in classification tasks to differentiate between groups. SSD-patients used less contingency (e.g., because) (p=.008) and multiclass connectives (e.g., as) (p<.001) than control participants. SSD-patients had higher minimum similarity of multiclass (adj-p=.04) and temporality connectives (e.g., after) (adj-p<.001), narrower similarity-range of expansion (e.g., and) (adj-p=.002) and multiclass connectives (adj-p=.04), and lower maximum similarity of expansion connectives (adj-p=.005). Using connectives' features alone, SSD-patients and controls could be distinguished with 85% accuracy. Our results show that SSD-speech can be distinguished from speech of control participants with high accuracy, based solely on connectives' features. We conclude that including connectives could strengthen computational models to categorize SSD.

Towards a temporospatial framework for measurements of disorganization in speech using semantic vectors.

Incoherent speech in schizophrenia has long been described as the mind making "leaps" of large distances between thoughts and ideas. Such a view seems intuitive, and for almost two decades, attempts to operationalize these conceptual "leaps" in spoken word meanings have used language-based embedding spaces. An embedding space represents meaning of words as numerical vectors where a greater proximity between word vectors represents more shared meaning. However, there are limitations with word vector-based operationalizations of coherence which can limit their appeal and utility in clinical practice. First, the use of esoteric word embeddings can be conceptually hard to grasp, and this is complicated by several different operationalizations of incoherent speech. This problem can be overcome by a better visualization of methods. Second, temporal information from the act of speaking has been largely neglected since models have been built using written text, yet speech is spoken in real time. This issue can be resolved by leveraging time stamped transcripts of speech. Third, contextual information - namely the situation of where something is spoken - has often only been inferred and never explicitly modeled. Addressing this situational issue opens up new possibilities for models with increased temporal resolution and contextual relevance. In this paper, direct visualizations of semantic distances are used to enable the inspection of examples of incoherent speech. Some common operationalizations of incoherence are illustrated, and suggestions are made for how temporal and spatial contextual information can be integrated in future implementations of measures of incoherence.

Sentence Length Effects on Intelligibility in Two Groups of Older Children With Neurodevelopmental Disorders.

The purpose of this study was to examine the impact of sentence length on intelligibility in two groups of older children with neurodevelopmental disabilities. Nine children diagnosed with cerebral palsy (CP) and eight children diagnosed with Down syndrome (DS), between the ages of 8 and 17 years, repeated sentences varying in length from two to seven words. Three hundred forty adult listeners (20 listeners per child) provided orthographic transcriptions of children's speech, which were used to calculate intelligibility scores. There was a significant main effect of sentence length on intelligibility for children with CP. Intelligibility significantly increased from two- and three-word sentences to four-, five-, and six-word sentences, then significantly decreased from four-, five-, and six-word sentences to seven-word sentences. There was a main effect of sentence length on intelligibility for children with DS. Intelligibility significantly increased from two-word sentences to four-, five-, and six-word sentences. The primary findings of this study include the following: (a) Unlike in typically developing children, sentence length continues to influence intelligibility well into adolescence for children with neurodevelopmental disorders, and (b) sentence length may influence intelligibility differently in children with CP than in children with DS; however, other factors besides the type of neurodevelopmental disorder (e.g., severity of speech motor involvement and/or cognitive-linguistic impairment) could play a role in the relationship between sentence length and intelligibility and must be investigated in future studies.

Noninvasive automatic detection of Alzheimer's disease from spontaneous speech: a review.

Alzheimer's disease (AD) is considered as one of the leading causes of death among people over the age of 70 that is characterized by memory degradation and language impairment. Due to language dysfunction observed in individuals with AD patients, the speech-based methods offer non-invasive, convenient, and cost-effective solutions for the automatic detection of AD. This paper systematically reviews the technologies to detect the onset of AD from spontaneous speech, including data collection, feature extraction and classification. First the paper formulates the task of automatic detection of AD and describes the process of data collection. Then, feature extractors from speech data and transcripts are reviewed, which mainly contains acoustic features from speech and linguistic features from text. Especially, general handcrafted features and deep embedding features are organized from different modalities. Additionally, this paper summarizes optimization strategies for AD detection systems. Finally, the paper addresses challenges related to data size, model explainability, reliability and multimodality fusion, and discusses potential research directions based on these challenges.

An analog-AI chip for energy-efficient speech recognition and transcription

Models of artificial intelligence (AI) that have billions of parameters can achieve high accuracy across a range of tasks1,2, but they exacerbate the poor energy efficiency of conventional general-purpose processors, such as graphics processing units or central processing units. Analog in-memory computing (analog-AI)3–7 can provide better energy efficiency by performing matrix–vector multiplications in parallel on ‘memory tiles’. However, analog-AI has yet to demonstrate software-equivalent (SWeq) accuracy on models that require many such tiles and efficient communication of neural-network activations between the tiles. Here we present an analog-AI chip that combines 35 million phase-change memory devices across 34 tiles, massively parallel inter-tile communication and analog, low-power peripheral circuitry that can achieve up to 12.4 tera-operations per second per watt (TOPS/W) chip-sustained performance. We demonstrate fully end-to-end SWeq accuracy for a small keyword-spotting network and near-SWeq accuracy on the much larger MLPerf8 recurrent neural-network transducer (RNNT), with more than 45 million weights mapped onto more than 140 million phase-change memory devices across five chips.

An enhanced method for dialect transcription via error‐correcting thesaurus

AbstractAutomatic speech recognition (ASR) has been widely used in the field of customer service, but the performance of general ASR in dialect transcription is not satisfactory, especially in Guangdong Province. Targeted training of ASR transcription engine will produce effect, but the training cost is high, and it is not suitable for small‐scale training with multiple dialects and frequencies. The complaint problems in the customer service field have obvious clustering and are suitable for few‐shot and multi‐frequency training. In view of this, in the actual engineering application, the method of ASR transcribed into the dialect error correction thesaurus is tried to be used to replace the wrong words, and have achieved good results. The optimization technology after automatic speech transcription proposed in this study can improve the recognition accuracy of general ASR by 13.75% for dialect words.

Assessment of the Perception and Worries of Saudi Healthcare Providers About the Application of Artificial Intelligence in Saudi Health Facilities.

Objective This study is aimed at assessing the perception and worries of Saudi healthcare providers about the application of artificial intelligence (AI) in Saudi healthcare facilities. Methods The study adopted a cross-sectional study involving 1026 Saudi healthcare providers between January 2023 and April 2023. The target population was healthcare providers across Saudi health facilities. Online questionnaires were administered through social media platforms. Data were analyzed using SPSS Statistics, version 26.0 (IBM Corp., Armonk, NY) to obtain important insights. Results The results of this study indicated that more than half (55.2%) of the respondents had good knowledge of AI, with (48.1%) of them being familiar with the application of AI in their specialty. A good proportion of the participants (57.9%) knew at least one term about the difference between machine learning and deep learning. More than half (69.9%) of the participants indicated that they had at one point in time used speech recognition or transcription application in their work. A large section (73.3%) of healthcare providers believed that AI would replace them at their job. A vast majority (84.9%) of the participants agreed that collaboration between medical schools with engineering and computer science faculties could be a game changer to provide a road for incorporating AI into medical curricula. The mean perception of AI in this study was 37.6 (SD=8.41; range 0-241). Age, level of health, health profession, and working experience all significantly impacted the positive perception score (p=0.021; p=0.031; p=0.041; p=0.026). However, there was no significant association between gender, nationality, and Saudi regions with a mean positive perception score. Conclusion There was a positive perception of AI among Saudi healthcare providers. Even though a substantial majority of Saudi healthcare providers were worried that AI would replace their jobs, the study revealed that AI serves as a crucial practitioner's tool rather than a physician's replacement.

Engaging Students’ Writing Competency on Metadiscourse Markers Using Donald Trump Presidential Speeches

This study aimed to investigate the implication of Donald Trump Presidential speeches as the teaching material of Metadiscourse Markers (MM) in writing class. It employed a qualitative study of English students after the pandemic situation. The research population consisted of 30 second-year students (male and female) from one selected class of Contextual Written Language Skills (CWLS) in the English Education Department of State University of Medan, Indonesia. The CWLS students’ writing ability was assessed using a questionnaire and Donald Trump Presidential speech transcripts taken from millercenter.org. The study referred to Hyland’s metadiscourse taxonomy (2005) and Lee’s contextual factors in writing (2016). The findings revealed that there was a distinct improvement in CWLS students’ writing ability between males and females on MM. This study also found that methodological variables such as teaching material and instructional design were some of the influencing factors in the learning process of MM using Donald Trump Presidential speeches.

ASSESSING THE EFFICACY OF LSTM, TRANSFORMER, AND RNN ARCHITECTURES IN TEXT SUMMARIZATION

The need for efficient and effective techniques for automatic text summarization has become increasingly critical with the exponential growth of textual data in different domains. Summarizing long texts into short summaries facilitates a quick understanding of the key information contained in the documents. In this paper, we evaluate various architectures for automatic text summarization using the TEDx dataset, a valuable resource consisting of a large collection of TED talks with rich and informative speech transcripts. Our research focuses on evaluating the performance of Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN) and Transformer architectures for automatic text summarization. We measure the accuracy of each model by comparing the generated summaries with human-written summaries. The findings show that the Transformer model achieves the highest accuracy, followed closely by the GRU model. However, LSTM, RNN exhibit relatively lower accuracies. We also investigate the trade-off between accuracy and conciseness in summarization. Our study reveals that the Transformer model succeeds in producing accurate and concise summaries, albeit at a higher computational cost. On the other hand, the GRU model strikes a desirable balance between accuracy and conciseness, making it a suitable choice. Overall, this research provides valuable insights into the effectiveness of different architectures for automatic text summarization and highlights the superiority of the Transformer and GRU models in this area.