If you are not fully satisfied with the default converter (Tika), you can create your own converter to import your document. If you have just created a project, do not import any document and configure your converter first, as below.
- Go to the Processing view
- Create a new converter
- Give a name to the converter
- Click on “Create from default configuration”
- Select the “OpenAI Speech to text” converter
- Select your model, language and save
- Activate converter
- Go to the main menu to import documents,
- You can now successfully select your audio files and import them.
- Upload your document
- Go to the document view to see the converted files
- To correct an audio transcription, it may be worthwhile using an additional model to re-punctuate and normalize the text.