CoolFace
Datasetpublic

Noothi/telugu-tech-indicf5-custom-voice

🎙️ Telugu Tech IndicF5 Custom Voice Dataset A 100% verified, clean, single-speaker Telugu Speech & Voice dataset specially formatted and phonetically cleaned for training and fine-tuning ai4bharat/IndicF5 and neural Text-to-Speech (TTS) models. All English technical terms, numbers, acronyms, and ASR mishearings have been converted into native Telugu phonetic script, cleaned of noise/brackets, and validated for optimal IndicF5 fine-tuning performance. 📊 Dataset… See the full description on the dataset page: https://huggingface.co/datasets/Noothi/telugu-tech-indicf5-custom-voice.

sourceHugging Facemitupdated 2mo agoView on Hugging Face
0likes66downloads
Dataset Card

🎙️ Telugu Tech IndicF5 Custom Voice Dataset

A 100% verified, clean, single-speaker Telugu Speech & Voice dataset specially formatted and phonetically cleaned for training and fine-tuning ai4bharat/IndicF5 and neural Text-to-Speech (TTS) models.

All English technical terms, numbers, acronyms, and ASR mishearings have been converted into native Telugu phonetic script, cleaned of noise/brackets, and validated for optimal IndicF5 fine-tuning performance.

📊 Dataset Statistics

  • —Total Clips: 455 audio files (.wav)
  • —Train Split: 437 clips (~70 minutes)
  • —Validation Split: 18 clips (~3 minutes)
  • —Total Audio Duration: 1 Hour 12 Minutes 48.5 Seconds (4,368.5 seconds)
  • —Total Dataset Size: ~1.20 GB
  • —Language: Telugu (te) with technical terms written in Telugu phonetic script
  • —Audio Quality: Clean isolated vocals (background music removed using Demucs htdemucs_ft)
  • —Transcriptions: 100% verified Telugu script phonetics, free of foreign artifacts & STT hallucinations.

📁 Source Videos & Breakdown

Video IDAudio ClipsDurationTopics Covered
-UMAP97L5q4113 clips19m 58.2sTechnical Architecture, CPU/IO Heavy Systems
RFAIqyP6mZc114 clips19m 50.5sMySQL, MyRocks Engine, Storage Optimization
kb2B9_A3TxE88 clips12m 40.8sMicroservices, Circuit Breaker, Exponential Jitter
5rTRRIwfKNo140 clips20m 18.8sSystem Design, Entity Layering, Uber Flow

💻 Usage with Hugging Face datasets

python
from datasets import load_dataset

# Load dataset directly from Hugging Face Hub
dataset = load_dataset("Noothi/telugu-tech-indicf5-custom-voice")

# Inspect sample entry
print(dataset['train'][0])

🛠️ Dataset Files & Structure

  • —metadata_clean.csv: Full cleaned metadata containing file_name,text
  • —train.csv: Training split for IndicF5 (437 rows)
  • —val.csv: Validation split for IndicF5 (18 rows)
  • —wavs/: Isolated 22050Hz mono WAV audio files