What's Changed

v0.5.0 by @R3gm in https://github.com/R3gm/SoniTranslate/pull/45

• Added option Overlap Reduction
• OpenAI API Key Integration for Transcription, translation, and TTS
• More output types: subtitles by speaker, separate audio sound, and video only with subtitles
• Access to a better-performing version of Whisper for transcribing speech on the Hugging Face Whisper page. Copy the repository ID and paste it into the 'Whisper ASR model' section in 'Advanced Settings'; e.g., kotoba-tech/kotoba-whisper-v1.1 for Japanese transcription available here
• Support for ASS subtitles and batch processing with subtitles
• Vocal enhancement before transcription
• Added CPU mode with app_rvc.py --cpu_mode
• TTS now supports up to 12 speakers
• OpenVoiceV2 integration for voice imitation
• PDF to videobook (displays images from the PDF)
• GUI language translation in Persian and Afrikaans
• New Language Support:
• Complete support: Estonian, Macedonian, Malay, Swahili, Afrikaans, Bosnian, Latin, Myanmar Burmese, Norwegian, Traditional Chinese, Assamese, Basque, Hausa, Haitian Creole, Armenian, Lao, Malagasy, Mongolian, Maltese, Punjabi, Pashto, Slovenian, Shona, Somali, Tajik, Turkmen, Tatar, Uzbek, and Yoruba
• Non-transcription: Aymara, Bambara, Cebuano, Chichewa, Divehi, Dogri, Ewe, Guarani, Iloko, Kinyarwanda, Krio, Kurdish, Kirghiz, Ganda, Maithili, Oriya, Oromo, Quechua, Samoan, Tigrinya, Tsonga, Akan, and Uighur

Full Changelog: https://github.com/R3gm/SoniTranslate/compare/0.4.0...0.5.0