Upload 3 files

Nepali Audio Dataset

1. Extended Description

This dataset contains 10 high-quality audio recordings collected by Dinesh Sirmal in Kathmandu, Nepal.
The dataset is designed for audio classification, sound recognition, and machine learning research.

2. Data Collection

- All audio files were recorded personally in various locations across Kathmandu to capture natural environmental sounds.
- Recordings were done using a standard microphone in WAV format for high fidelity.
- Audio clips include a variety of everyday sounds such as:
- Human speech
- Vehicles Noise
- Environmental noises (rain, traffic, alarms, footsteps)
- Each clip is labeled carefully and stored in a labels.csv file for easy use in supervised learning tasks.

3. Purpose

The dataset is intended for:
1. Audio classification experiments – training machine learning models to classify sound types, accent detectation.
2. Multi-modal projects*– combining audio with other datasets (video, text) for advanced AI research.
3. Educational purposes – for students or researchers learning sound recognition.

4. Structure

- audio/ → 10 WAV audio files (~10 MB each)
- labels.csv → file names with corresponding labels
- README.md → description, usage, and license information

5. License and Citation

- License: CC BY 4.0 – Users must give credit to Dinesh Sirmal when using this dataset.
- Citation:

--PJHxphWEs_30000.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fba12667a61798c00d54663a354c758e4ab37132d5d3ba8f9625803f65902a53
+size 960044

--aaILOrkII_200000.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d27ad192477b086982e8a001abca7ee35fbd156cc6a56e4a5f185c4f2392dc8d
+size 960044

--ekDLDTUXA_30000.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2737ff907ca66ab3e078c9a9b191ac06a966bde871b5d19716daf35cf9a8565
+size 960044