This dataset is a gender-specific subset of the original DailyTalk TTS dataset.It contains English conversational speech paired with text transcripts, filtered and separated by speaker gender.
This version includes:
Two columns:
text: transcription
audio: 24 kHz speech waveform
One split (train) with 11,906 samples
No audio processing or text modifications were performed. The dataset is a structured subset of the original source.
Dataset Details… See the full description on the dataset page: https://huggingface.co/datasets/innovationm-ai/dailytalk-female.
