IndicTTS Hindi: High-Quality Hindi Speech Recordings for Text-to-Speech

Approximately 10.33 hours of high-quality Hindi speech recordings from the Indic TTS Database project, contributed by SPRINGLab. The dataset contains audio from both male and female speakers, with corresponding text transcriptions. It was last updated on November 5, 2024.

Use Cases

• Training text-to-speech models based on high-quality Hindi audio recordings.
• Fine-tuning speech synthesis models based on speaker-specific recordings (male/female).
• Developing multilingual TTS systems based on data from the Indic TTS project.
• Researching speech prosody and pronunciation for Hindi based on aligned audio and text.

Strengths

• Contains approximately 10.33 hours of audio, with balanced male (5.16 hours) and female (5.18 hours) speaker contributions.
• High-quality WAV format audio recorded at a 48kHz sampling rate.
• Includes corresponding text transcriptions, which are essential for TTS model training.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and dataset size are unknown, which may limit suitability assessment.
• Data may reflect source bias inherent to the specific Indic TTS Database project collection.

Provenance

Source

SPRINGLab, derived from the Indic TTS Database project.

Collection Method

Monolingual recordings from the Indic TTS Database project.

Time Range

null

Freshness

Last updated 2024-11-05 10:14:34; freshness should be verified.

Geography

null