eLearning Content Creation
Emotion Tags
normalslowcryingsleepysighchuckle
Guided Emotion and Intonation
Input Streaming for Lower Latency
Language/Country Support
EnglishChineseSpanishTurkishGermanArabicFrenchHindiAfrikaansArmenianAzerbaijaniBulgarianCroatianCzechDanishDutchEstonianFinnishGreekHebrewHungarianIcelandicIndonesianItalianJapaneseKazakhKoreanLatvianLithuanianMalayNorwegianPersianPolishPortugueseRomanianRussianSerbianSlovakSlovenianSwedishTamilThaiUkrainian
LLM-based Customizability
Model Tokenizer Type
Non-streaming (CNN-based) tokenizer
Open Source Release Planned
Orpheus Speech Models
Medium (3B)Small (1B)Tiny (400M)Nano (150M)
Pretrained and Finetuned Models
Pretrained modelsFinetuned models
Processing Speed
3-5 minutes
Sample Finetuning Scripts
SEO and Copywriting Tools
Sliding Window Detokenizer
Smart Transcription Accuracy
98%
Streaming Inference Speed
Faster than playback on A100 40GB for 3B model
Supported Content Types
Twitter threadsBlog postsNewslettersInteractive quizzesLearning materials
Training Data Volume
100k+ hours of speech, billions of text tokens