Upload utt_spk_text_pure.tsv

In this TSV file, I have converted Nepali numeric values to their corresponding Nepali text, using a mapping such as:
{
"०": "शून्य",
"१": "एक",
"२": "दुई",
"३": "तीन",
"४": "चार",
"५": "पाँच",
"६": "छ",
"७": "सात",
"८": "आठ",
"९": "नौ"
}
Additionally, since the audio files were trimmed to remove silence at the beginning and end, some files ended up containing no speech. These silent or empty audio files were identified and manually removed.
Most of the data cleaning tasks, including text correction and file verification, were performed manually.

data/utt_spk_text_pure.tsv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33b6d8aff021b8844e0f6376dc0cf2d6ff14b71652ae4c23b4c534e6854afca9
+size 12159956