|
--- |
|
license: openrail |
|
task_categories: |
|
- summarization |
|
language: |
|
- en |
|
tags: |
|
- Ratan |
|
- Tata |
|
- tts |
|
- voicecloning |
|
pretty_name: S |
|
size_categories: |
|
- n<1K |
|
--- |
|
This dataset consists of audio samples and metadata for a voice cloning task, particularly designed for the TTS (Text-to-Speech) system. The dataset has been pre-processed and includes high-quality audio files featuring the voice of Ratan Tata in English. |
|
Dataset Information |
|
|
|
License: OpenRAIL (Open Responsible AI License) |
|
Total Dataset Duration: 2880.55 seconds (approx. 48 minutes) |
|
Minimum Audio Length: 3.65 seconds |
|
Maximum Audio Length: 11.77 seconds |
|
Average Audio Length: Varies, with an average around 7 seconds per clip. |
|
|
|
Audio Format |
|
|
|
Audio Channels: Mono (1 channel) |
|
Sample Rate: 16 kHz |
|
Precision: 16-bit |
|
Audio Encoding: 16-bit Signed Integer PCM |
|
Typical Bit Rate: 256 kbps |
|
|
|
Example of an audio file specification (using SoX tool): |
|
|
|
Input File : 'converted_ratan_tata_tts_1.wav' |
|
Channels : 1 |
|
Sample Rate : 16000 |
|
Precision : 16-bit |
|
Duration : 00:00:07.10 (approx. 7 seconds) |
|
File Size : 227 KB |
|
Bit Rate : 256 kbps |
|
Sample Encoding: 16-bit Signed Integer PCM |
