Multilingual VIDEO Dataset with Speaker Stems, Diarized Transcripts, Scene Descriptions | 2.8M+ ...

The ACNetwork Multilingual Conversational Media Dataset delivers more than 2.8M hours of studio and UCG audio and video data across 133 languages: including English, Portuguese, Spanish, Russian, French, German, Italian, Japanese, Korean and more.

Each file features high-fidelity audio with isolated stems, transcripts at word and utterance level, speaker diarization, emotion and tone labels, scene descriptions, related metadata, and a human-verified subset for benchmarking. All content is fully rights-cleared and indemnified for AI/LLM training and use cases, making it ideal for speech-to-text, multimodal retrieval, video generation, cross-lingual fine-tuning, and emotion classification.

The dataset spans sports, news, entertainment, gaming, true crime, and more, with new content added monthly to ensure domain diversity and freshness. Delivery is via Google Drive or AWS S3, and data is provided in multiple formats (JSON, SRT, VTT, TXT).

This is the definitive global multilingual media dataset for enterprise-scale AI and LLM training - combining audio and video sources at unmatched depth and breadth. Direct licensing by inquiry only - contact ACNetwork for commercial terms or expanded sample access.