TTS

This dataset was recorded by a 30-year-old voice actor with a friendly and soft timbre.

【Features】 Multi-domain: 4 domains Couple conversation, E-commerce live streaming, Declarative tone, Free conversation Multi-emotion: 14 emotions Pleasure, Dissatisfaction, Fear, Tenderness, Lyricism, Sadness, Sternness, Friendliness, Whispering, Apology, Excitement, Coquetry, Anger, Calmness

This dataset was recorded by a 30-year-old voice actor with a friendly and soft timbre.

Natural Style: Topic-expanding casual conversation, e.g., daily life, hobbies, childhood experiences, with continuous context. Multi-Emotion: Comforting, apologetic, happy, fearful, coquettish, sad, angry, joyful, disgusted, serious. Rich Corpus: Includes cold jokes, cheesy pick-up lines, idioms/proverbs, Chinese characters, poetry, novel texts, and mixed Chinese-English content.

【Corpus】Centered around character settings, single sentences 【Emotion】Neutral, happiness, anger, sadness, fear, disdain, concern, seriousness, inner thoughts 【Labeling】Pronunciation, prosody, paralanguage 【Paralanguage】Crying, stuttering, laughing, snorting, sighing, smacking lips, coughing, breathing sounds, and other paralinguistic phenomena

【Corpus】Centered around character settings, single sentences 【Emotion】Neutral, happiness, anger, sadness, fear, disdain, concern, seriousness, inner thoughts 【Labeling】Pronunciation, prosody, paralanguage 【Paralanguage】Crying, stuttering, laughing, snorting, sighing, smacking lips, coughing, breathing sounds, and other paralinguistic phenomena

Recording Method: Two speakers per session were recorded in the same studio, with each individual's voice captured separately as a single continuous audio track.

This dataset was recorded by a 7-year-old girl with a clear and sweet voice.