Dataset Viewer
Auto-converted to Parquet Duplicate
AudioID
stringclasses
1 value
audio_text
stringclasses
1 value
gender
stringclasses
1 value
duration
stringclasses
1 value
format
stringclasses
1 value
__index_level_0__
stringclasses
1 value
__index_level_1__
stringclasses
1 value
prirodan ritam i dosljedan kvalitet zvuka za primjene mašinskog učenja. Sistemi umjetne inteligencije oslanjaju se na velike količine visokokvalitetnih glasovnih podataka kako bi poboljšali svoje performanse. Snimanjem ovih rečenica jasno i prirodno
doprinosite razvoju naprednije tehnologije prepoznavanja govora.
male
0:36
Mp3
Bosnian
Ovo je glasovni audio skup podataka koji možete koristiti za potrebe treniranja umjetne inteligencije. Svaki zapis je pažljivo pripremljen kako bi se osigurala jasna izgovor

🎧 Bosnian Speech Dataset

The Bosnian Speech Dataset is a high-quality speech audio dataset developed to provide reliable and scalable audio data for modern AI and machine learning applications. It includes 169 hours of audio data across 798 files, delivered in MP3 and WAV formats, with a total size of 371 MB. This well-structured audio dataset ensures balanced and diverse voice data, with 51% female and 49% male speakers, and a wide age distribution from 18 to 50+ years. The dataset language is Bosnian, with speakers from Bosnia and Herzegovina, Serbia, Montenegro, and Croatia, making it a representative language speech dataset for regional linguistic variation.


🔗 Learn more:
https://speech-data.ai/datasets/bosnian/


📊 Dataset Metadata

Field Value
📜 License CC BY-NC-ND 4.0
🎯 Task Categories Automatic Speech Recognition
🌍 Language Bosnian (bs)
🏷️ Tags Audio, Speech, Machine, Machine Learning, Speech Recognition, Data
📦 Size Category n < 1K

🚀 Use Cases

This Bosnian speech dataset supports a wide range of AI applications, including speech recognition, voice assistant development, language identification, and translation systems. The structured speech data enables efficient training of acoustic and language models, making it a reliable speech recognition dataset for both research and production use. It is also suitable for accent recognition and multilingual AI workflows requiring high-quality voice data.


⭐ Key Value

The core value of this speech dataset lies in its regional diversity, balanced demographics, and production-ready format. It delivers consistent and high-quality audio data, enabling the development of robust and scalable voice-enabled systems. This voice dataset helps improve model performance in real-world multilingual and cross-accent environments.

Downloads last month
9