Speech Recognition Datasets

This dataset contains 1,013 hours of Brazilian Portuguese conversational speech, featuring natural conversations and monologues recorded by native speakers from diverse regions. The recordings reflect real-world speaking styles and provide rich linguistic diversity for speech AI development. Each recording includes a transcript, speaker ID, gender, and supporting metadata, enhancing model performance in real and complex tasks. The dataset has undergone quality validation by multiple AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, and PIPL compliant.

This speech dataset mirrors real-world interactions. Each audio is transcribed and includes text content, speaker ID, gender, age, accent, and other attributes. Our dataset collected from a wide and diverse range of speakers (children aged 12 and under), improves the model's performance on real and complex tasks by geographic location. Quality tested by various AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, PIPL complied.

This dataset contains 406 hours of European Portuguese spontaneous dialogue speech collected through topic-based conversations covering more than 20 domains. Each recording includes accurate transcripts, speaker ID, gender, age, and additional metadata. The dataset was collected from 590 native European Portuguese speakers across diverse regions, enhancing model performance in real and complex tasks. Quality tested by various AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, PIPL complied.

This 101 hours Italian children speech dataset reflects real-world interactions. Each audio sample is transcribed and includes rich metadata such as speaker ID, gender, age, accent, timestamps, and noise-related attributes. Our dataset is collected from a broad and diverse group of speakers (children aged 12 and under) with wide geographical distribution, thus improving the model's performance on real-world, complex tasks. Quality tested by various AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, PIPL complied.

Malay(Malaysia) Spontaneous Dialogue Smartphone speech dataset, collected from dialogues based on given topics, covering 20+ domains. Transcribed with text content, speaker's ID, gender, age and other attributes. Our dataset was collected from extensive and diversify speakers(228 native speakers), geographicly speaking, enhancing model performance in real and complex tasks. Quality tested by various AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, PIPL complied.

This 162 hours of French children speech dataset mirrors real-world interactions. Each audio sample is transcribed and includes text content, speaker ID, gender, age, accent, and other attributes. Our dataset collected from a wide and diverse range of speakers (children aged 12 and under), improves the model's performance on real and complex tasks by geographic location. Quality tested by various AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, PIPL complied.

This dataset contains 1003 hours of Hindi speech audio, mirrors real-world interactions. Each utterance is transcribed with text content, speaker's ID, gender, and other attributes. Our dataset was collected from extensive and diversify speakers, geographicly speaking, enhancing model performance in real and complex tasks. Quality tested by various AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, PIPL complied.

This Korean Telephony Speech Dataset contains 136 hours of spontaneous dialogue recorded over phone calls. Covering over 20 real-life domains including customer service, e-commerce, finance, travel, and daily conversations, the dataset features natural two-speaker conversations collected via diverse telephony channels. Each sample is transcribed and annotated with speaker ID, gender, age, and other metadata. Data was collected from 216 native Korean speakers across different regions, enhancing model generalization. Ideal for automatic speech recognition (ASR), speaker diarization, and call center conversational AI systems. All data complies with GDPR, CCPA, and PIPL for responsible and legal AI development.

Indonesian(Indonesia) Spontaneous Dialogue Smartphone speech dataset, collected from dialogues based on given topics, covering 20+ domains. Transcribed with text content, speaker's ID, gender, age and other attributes. Our dataset was collected from extensive and diversify speakers(412 native speakers), geographicly speaking, enhancing model performance in real and complex tasks. Quality tested by various AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, PIPL complied.