Iban Language Speech Corpus for ASR Training

Audio files and corresponding transcriptions for training Automatic Speech Recognition models for the Iban language. The dataset was created by user 'meisin123' and was last updated in November 2023. It is hosted on the Hugging Face platform.

Use Cases

• Train an ASR model to transcribe Iban audio files into text transcriptions.
• Fine-tune a pre-trained multilingual speech model using Iban audio and its corresponding text labels.
• Analyze acoustic features from the provided wav files for Iban phoneme or speech pattern research.

Strengths

• Contains aligned audio and text data, which is essential for supervised ASR training.
• Dataset was updated in 2023, indicating recent activity.

Limitations

• The total number of audio files, hours of speech, and speaker demographics are unknown.
• No information on audio quality, recording conditions, or potential class imbalance in vocabulary.

Provenance

Source

Original dataset repository referenced in the description, but not named.

Collection Method

Collection method is not specified in the provided description.

Freshness

Last updated on 2023-11-02.