Launched by SubEasy.aiAI subtitle and speech-to-text platform service of the same nameIt claims to be able to convert sounds with high accuracy, or correctly identify the content of the speaker's statement and convert it into text content, which can then be used for subsequent content translation, summarization and other application needs.
Based on OpenAI's Whisper model, with customized optimization
The SubEasy AI subtitle and speech-to-text platform service is based on OpenAI's Whisper model and incorporates SubEasy.ai customization and optimization, allowing audio content to be converted to text with higher accuracy. The "Clean Recognition" function can even remove ambient noise from the audio content before conversion, making the text recognition results more accurate.
Currently, SubEasy supports audio sources including MP3, MP4, M4A, MOV, AAC, WAV, OGG, OPUS, MPEG, WMA and other formats, and can even correspond to the sound of YouTube video content.
SubEasy can also automatically identify and annotate individual speakers in conference audio content, making subsequent text conversion easier to read and organize. Furthermore, the conversion results can be translated into up to 100 languages, including English, Japanese, Chinese, Thai, Korean, Russian, Spanish, and even Cantonese, thereby reducing misunderstandings in international conferences.
Proprietary translation results can be set for professional fields and specific words
SubEasy also incorporates a translation glossary feature, allowing users to set specific translation results for specific words, ensuring more accurate translations for technical terms and industry jargon in different fields. The summary function, combined with artificial intelligence technology, analyzes and summarizes the voice-to-text conversion results, quickly summarizing them for easier reading. Users can even "ask" questions about the converted text through the dialogue function, with AI providing answers based on their understanding.
Simplified editing interface after conversion
Users can directly switch to the "Transcript View" on the SubEasy service platform to organize the converted text content in a comprehensive manner, manually edit and mark key points, and use intelligent reflow technology to readjust sentence segments according to the context to ensure a good viewing and reading experience. After editing, files can be exported in various file formats such as VTT, Word, Text, MD, LRC, JSON, ASS, CSV, STL, and PDF.
Suitable for online meetings, multinational team collaboration, online learning, and content creation
SubEasy's applications include recording corporate online meeting content, multinational team collaboration, or students taking notes on online learning content. It can even reduce the difficulty of subtitle production for content creators.
SubEasy.ai Product Manager Terry Gu said, "We understand that Taiwanese users rely heavily on the accuracy and convenience of text transcription for remote conferencing, learning, and cross-language communication. We will continue to optimize SubEasy's audio-to-text feature to help businesses and individuals quickly grasp key points and consolidate knowledge. Going forward, we will also expand the application of AI in everyday life, enabling more people to easily record and learn even in their busy lives."
Provides basic free use, and also provides pay-as-you-go plans for occasional use
Currently, SubEasy offers a basic free usage model, which allows users to convert up to 3 minutes of audio content three times a day for free. It can be used through the basic conversion channel and can also be translated using artificial intelligence models. It also provides subtitle rearrangement, interactive conversations through ChatGPT, and video compression and export functions.
The monthly or annual subscription plans offer more features, longer content, and fewer restrictions. Exported videos also do not include watermarks. Enterprise users can customize features based on their needs, and a pay-as-you-go plan is available for occasional use.
