You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
What I want to discuss is when using [--initial_prompt], what format of example sentences will you use to guide the generation of correct content?
For example, the audio content is about painting teaching, and there will be some professional words in it. During recognition, it will be confused with some homophones, or the pronunciation will be ambiguous, resulting in some words in the generated results that do not conform to the overall content or are completely different words. This situation is rare in English and can basically be corrected by the translation machine during translation.
But in the case of Japanese and Korean, homophones or words with similar pronunciations appear more often. For example, [paper] and [hair] in Japanese.
I have also read [openai/whisper] about how to use examples and questions and answers. When someone uses example sentences, it will guide the accuracy of the generated content, but this is not 100% true. So I want to ask you what format of example sentences can guide the generation of correct content.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
What I want to discuss is when using [--initial_prompt], what format of example sentences will you use to guide the generation of correct content?
For example, the audio content is about painting teaching, and there will be some professional words in it. During recognition, it will be confused with some homophones, or the pronunciation will be ambiguous, resulting in some words in the generated results that do not conform to the overall content or are completely different words. This situation is rare in English and can basically be corrected by the translation machine during translation.
But in the case of Japanese and Korean, homophones or words with similar pronunciations appear more often. For example, [paper] and [hair] in Japanese.
I have also read [openai/whisper] about how to use examples and questions and answers. When someone uses example sentences, it will guide the accuracy of the generated content, but this is not 100% true. So I want to ask you what format of example sentences can guide the generation of correct content.
Beta Was this translation helpful? Give feedback.
All reactions