DuBLaB-en-fr-0.7 is a bilingual dataset of English-French audio segments, aligned and extracted from audiovisual sources. This dataset is designed for automatic speech recognition (ASR), machine translation, cross-lingual style mapping, and text-to-speech (TTS) applications.