ASR Datasets
Collection
Datasets for training and evaluation of ASR models. • 5 items • Updated
Datasets:
Please provide the required information and accept the Terms and Conditions to request access to the dataset
Terms and Conditions
This dataset may be used solely for research purposes and for developing artificial intelligence tools focused on linguistic objectives. It can be accessed by any entity collaborating on projects promoted by the Centro Ramón Piñeiro de Investigación en Humanidades and Universidade de Santiago de Compostela, particularly those involved in the implementation of the Nós project. Dissemination of the voice recordings in an open-access manner or their public exposure is strictly prohibited. By accessing and using this dataset, you agree to comply with all applicable laws and ethical standards regarding the protection of individual rights. Users are strictly prohibited from using the dataset in any way that infringes upon the rights, privacy, or dignity of any individual represented within it. Any misuse, including but not limited to attempts to engage in discriminatory, harmful, or unlawful activities, is expressly forbidden.
This document is a legal contract between you and the dataset provider. By accessing and using the dataset, you acknowledge and agree to abide by the Terms and Conditions.
Our team may take a few days to process your access request. Thank you in advance for your patience.
Log in or Sign Up to review the conditions and access this dataset content.
Nos_ImosNavegando-GL
Dataset description
Nos_ImosNavegando-GL is a Galician dataset designed for the evaluation and/or development of automatic speech recognition (ASR) systems. The corpus is based on television content from the entertainment genre, specifically the programme Imos Navegando.
The dataset contains approximately 14 hours of manually reviewed text-aligned audio. The review process involved checking and correcting the available text, adapting it to the actual audio recording, and revising the temporal alignment between the audio and the transcription.
Out of the 14 hours:
- 10 hours initially included human subtitles previously aligned with the audio.
- 4 hours initially included automatic transcriptions previously aligned with the audio.
The work carried out focused on the manual revision of these materials in order to improve both the textual quality and the temporal alignment of the transcriptions, producing a resource suitable for ASR evaluation in Galician.
Data Type
The corpus includes audio from a Galician television entertainment programme and time-aligned transcriptions. The nature of the audiovisual content makes it possible to capture speech in a media-related setting, with features such as spontaneous or semi-spontaneous interaction, prosodic variation, speaker changes, background music or noise, and other elements typical of television production.
Review and Alignment Process
The manual review was carried out by trained linguists using ELAN, a tool for the annotation and temporal alignment of audiovisual data. The process included:
reviewing the available text; correcting the transcription when it did not match the audio; adapting the textual segmentation to the recording; revising and adjusting the temporal alignment between audio and text; checking the overall consistency of the resulting segments.
Intended Uses
This dataset can be used for:
- evaluation of ASR systems for Galician;
- evaluation of ASR performance on television speech;
- study of speech phenomena in entertainment television programmes;
- development of linguistic and technological resources for Galician, etc.
Limitations
The corpus comes from a specific domain, namely television entertainment programmes. Therefore, results obtained on this dataset should not necessarily be interpreted as representative of all spoken registers of Galician.
Funding and acknowledgements
This work is funded by the Ministerio para la Transformación Digital y de la Función Pública - Funded by EU – NextGenerationEU within the framework of the project Desarrollo de Modelos ALIA. (Esta publicación del proyecto Desarrollo de Modelos ALIA está financiada por el Ministerio para la Transformación Digital y de la Función Pública y por el Plan de Recuperación, Transformación y Resiliencia – Financiado por la Unión Europea – NextGenerationEU).
We would like to thank the Galician Public Television and Radio Corporation (CRTVG) for their kind collaboration in providing the original data.
Our gratitude also to the reviewers of this dataset, Reverso Comunicación.
For more information, please go to nos.gal or contact the Nós project at proxecto.nos@usc.gal.
Licensing Information
Creative Commons Attribution 4.0 International
Terms and Conditions
By accessing and using this dataset, you agree to comply with all applicable laws and ethical standards regarding the protection of individual rights. The dataset contains voice files, transcripts, and metadata, including participant identity information, provided solely for research and development purposes. Users are strictly prohibited from using the dataset in any way that infringes upon the rights, privacy, or dignity of any individual represented within it. Any misuse, including but not limited to attempts to engage in discriminatory, harmful, or unlawful activities, is expressly forbidden. Dissemination of the voice recordings in an open-access manner or their public exposure is strictly prohibited.
Citation information
If you use this dataset, please cite it as:
Proxecto Nos. 2026. Nos_ImosNavegando-GL. URL: https://huggingface.co/datasets/proxectonos/Nos_ImosNavegando-GL.
- Downloads last month
- 16