With the rapid development of artificial intelligence and machine learning technologies, automatic speech recognition (ASR) and text-to-speech (TTS) have become key components of the digital transformation of society. The Kazakh language, as a representative of the Turkic language family, remains a low-resource language with limited audio corpora, language models, and high-quality speech synthesis systems. This study provides a comprehensive analysis of existing speech recognition and synthesis models, emphasizing their applicability and adaptation to the Kazakh language. Special attention is given to linguistic and technical barriers, including the agglutinative structure, rich vowel system, and phonemic variability. Both open-source and commercial solutions were evaluated, including Whisper, GPT-4 Transcribe, ElevenLabs, OpenAI TTS, Voiser, KazakhTTS2, and TurkicTTS. Speech recognition systems were assessed using BLEU, WER, TER, chrF, and COMET, while speech synthesis was evaluated with MCD, PESQ, STOI, and DNSMOS, thus covering both lexical–semantic and acoustic–perceptual characteristics. The results demonstrate that, for speech-to-text (STT), the strongest performance was achieved by Soyle on domain-specific data (BLEU 74.93, WER 18.61), while Voiser showed balanced accuracy (WER 40.65–37.11, chrF 80.88–84.51) and GPT-4 Transcribe achieved robust semantic preservation (COMET up to 1.02). In contrast, Whisper performed weakest (WER 77.10, BLEU 13.22), requiring further adaptation for Kazakh. For text-to-speech (TTS), KazakhTTS2 delivered the most natural perceptual quality (DNSMOS 8.79–8.96), while OpenAI TTS achieved the best spectral accuracy (MCD 123.44–117.11, PESQ 1.14). TurkicTTS offered reliable intelligibility (STOI 0.15, PESQ 1.16), and ElevenLabs produced natural but less spectrally accurate speech.

This study presents the application of free, open-source artificial intelligence (AI) techniques to advance machine translation for low-resource Turkic languages such as Kazakh, Azerbaijani, Kyrgyz, Turkish, Turkmen, and Uzbek. This machine translation problem for Turkic languages is part of a project to generate meeting minutes from speech transcripts. Due to limited parallel corpora and underdeveloped linguistic tools for these languages, traditional machine translation approaches often underperform. The goal is to reduce digital inequality for these languages and to support scalability. We investigate the effectiveness of free open-source pre-trained specialized and general-purpose AI models for morphologically rich state Turkic languages. This research includes developing parallel corpora for six Turkic languages, fine-tuning, and performance evaluation using BLEU, WER, TER, and chrF metrics. The parallel corpora for five pair languages, each of 300,000 and 500,000 sentences, were generated and cleaned. The results for corpora 500,000 parallel sentences show significant improvements compared with baseline NLLB-200 1.3B on average: BLEU increased by 23.81 points, chrF increased by 26.05 points, and WER and TER decreased by 0.36 and 33.95, respectively, after cleaning and fine-tuning. Six Turkic-language multilingual parallel corpora of 3 885 542 sentences were developed and the fine-tuning of NLLB-200 1.3B shows the following, compared with the results for 500,000 cleaned corpus: BLEU increased by 4.3 points, chrF increased by 1.7 points, and WER and TER decreased by 0.1 and 4.75, respectively These results demonstrate the high efficiency of corpus cleaning and synthetic data generation to improve the quality of machine translation for low-resource Turkic languages using AI models. These results were confirmed by external evaluation on the FLORES 200 dataset and human evaluation. The scientific contribution of this article is the development of a methodology for generating parallel corpora using a specialized AI model of machine translation and fine-tuning the specialized AI model on the created corpora, creating new multilingual parallel corpora of Azerbaijan–Kazakh, Kyrgyz–Kazakh, Turkish–Kazakh, Turkmen–Kazakh, and Uzbek–Kazakh pairs using the proposed methodology, cleaning them, and conducting fine-tuning experiments.

The development of parallel audio corpora for Turkic languages, such as Kazakh, Uzbek, and Tatar, remains a significant challenge in the development of multilingual speech synthesis, recognition systems, and machine translation. These languages are low-resource in speech technologies, lacking sufficiently large audio datasets with aligned transcriptions that are crucial for modern recognition, synthesis, and understanding systems. This article presents the development and experimental evaluation of a speech corpus focused on Turkic languages, intended for use in speech synthesis and automatic translation tasks. The primary objective is to create parallel audio corpora using a cascade generation method, which combines artificial intelligence and text-to-speech (TTS) technologies to generate both audio and text, and to evaluate the quality and suitability of the generated data. To evaluate the quality of synthesized speech, metrics measuring naturalness, intonation, expressiveness, and linguistic adequacy were applied. As a result, a multimodal (Kazakh–Turkish, Kazakh–Tatar, Kazakh–Uzbek) corpus was created, combining high-quality natural Kazakh audio with transcription and translation, along with synthetic audio in Turkish, Tatar, and Uzbek. These corpora offer a unique resource for speech and text processing research, enabling the integration of ASR, MT, TTS, and speech-to-speech translation (STS).

With the rapid development of artificial intelligence and machine learning technologies, automatic speech recognition (ASR) and Text-to-Speech (TTS) are becoming key components of the digital transformation of society. The Kazakh language, a representative of the Turkic language family, is a low-resource language. The analysis shows significant limitations in the availability of key components of the language infrastructure, namely audio corpora, language models, and high-quality speech synthesis systems. However, for the Kazakh language, which has unique phonetic, morphological, and syntactic features, the level of development of ASR/TTS technologies still lags significantly behind their counterparts for widely spoken languages. This study aims to comprehensively analyze existing speech recognition and text-to-speech models and platforms, emphasizing their applicability and adaptation to the Kazakh language. Particular attention is paid to the linguistic and technical barriers that hinder the effective integration of the Kazakh language into modern voice technologies, including the agglutinative structure, rich vowel system, and phonemic variability. This study aims to comprehensively evaluate a diverse range of existing Speech-to-Text (STT) and TTS models and platforms in terms of their applicability to the Kazakh language. We have tested both open and commercial solutions, including Whisper, GPT-4 Transcribe, ElevenLabs, OpenAI TTS, Voiser, KazakhTTS2, TurkicTTS, and others. The assessment of speech recognition was based on the WER, TER, chrF, BLEU, and COMET metrics. In contrast, speech synthesis was evaluated using MCD, PESQ, STOI, and DNSMOS, which cover both lexical-semantic and acoustic-perceptual characteristics. Based on the analysis, it has selected the most accurate universal STT system model, not trained on local data, which demonstrated high accuracy and semantic proximity. In the field, we have identified a model that combines minimal spectral distortions with high subjective sound quality.

This study addresses the critical issue of data scarcity for low-resource languages, particularly focusing on the methodology for creating a corpus of parallel sequences in two low-resource languages. The lack of large-scale, high-quality bilingual datasets significantly hinders the development of neural machine translation systems for such languages. In this work, a comparative analysis of AI systems for generating parallel corpus on a test dataset is conducted, with selection criteria based on accessibility (free to use), translation quality, and efficiency. AI system was selected based on predefined criteria, and its performance in generating parallel data was assessed. As an example, a sizable Kyrgyz-Kazakh parallel corpus was created. However, error analysis revealed that approximately 0.5% of the translations contained inaccuracies, highlighting the need for further post-editing and model refinement. This study contributes to the advancement of resource development for low-resource language pairs and provides practical insights into the efficient creation of parallel corpus using modern AI systems.