Refine
Year of publication
- 2021 (1)
Document Type
- Master's Thesis (1)
Language
- German (1)
Has Fulltext
- yes (1)
Keywords
Faculty
More and more often, spoken information must and should be available in written form. For this purpose, various transcription programs try to support the user with various conveniences when transcribing the source material. A variety of online services go one step further and provide a ready-to-use, automatically generated transcription for a fee. Since the fees can be very expensive for the individual user and the online services may not always be used for privacy reasons, the goal of this work is to implement an open offline alternative. This alternative should be an open source editor based on the open speech-to-text-engine DeepSpeech and should on one hand provide the user with an offline transcription and on the other hand support him in correcting it. To achieve this goal, first the traditional speech recognition and eventually DeepSpeech will be described. This is followed by the conception and implementation of the editor. Since this project is explicitly intended to be an open source project, the last part will take a closer look at the release.