unified_multilingual_dataset_of_emotional_human_utterances and multilingual_speech_valence_classification_datasets
These are complementary datasets, with A providing a unified dataset of emotional utterances that could be enriched or expanded by B's multilingual datasets with raw audio specifically for valence classification, to build a more comprehensive speech emotion recognition system.
About unified_multilingual_dataset_of_emotional_human_utterances
michen00/unified_multilingual_dataset_of_emotional_human_utterances
A unified dataset of multilingual emotional human utterances
This dataset provides a collection of over 87,000 audio files of human speech and vocalizations, each labeled with the speaker's gender, emotional category (like positive, negative, neutral), and unique speaker ID. It consolidates diverse sources, offering recordings in English and eight other languages. This resource is designed for researchers and practitioners in fields like human-computer interaction, mental health, or customer experience analysis who need categorized vocal data.
About multilingual_speech_valence_classification_datasets
michen00/multilingual_speech_valence_classification_datasets
Multilingual datasets with raw audio for speech emotion recognition
This project offers a collection of diverse datasets featuring raw audio recordings of human speech and vocalizations. Its purpose is to provide resources for researchers and developers building conversational AI systems or speech analytics tools that can understand and respond to human emotions. The datasets include audio samples in various languages, along with labels indicating emotional valence (e.g., positive, negative, neutral).
Scores updated daily from GitHub, PyPI, and npm data. How scores work