bengali-tts-missing-v1

Bengali TTS — Missing Rows This dataset contains the rows from rwd51/bengali-tts-combined that are not present in the filtered repository dipit099/bengali-tts-filtered-v3. In other words: source samples that were never scored by Whisper/Gemini in the original transcription pipeline and therefore never made it into the quality-filtered output. Source rwd51/bengali-tts-combined (397,437 rows, 780 parquet shards) dipit099/bengali-tts-filtered-v3 (326,647 KEEP rows)… See the full description on the dataset page:

Source: Hugging Face Hub (smam/bengali-tts-missing-v1). Metadata imported from the dataset’s Hub tags.