Vilas Ninawe
Vilas Ninawe
Please check whether you are using correct vocab file or not. For multilingual model, filters_vocab_multilingual.bin is used. And for english only model, filters_vocab_gen.bin is used. Modify isMultilingual variable as per...
@arunesh try with and without multilingual by changing isMultilingual variable. If it still doesn't work, share model path.
@AbelAditya can you please share what is expected output and what is actual output? If possible, please share audio file.
@AbelAditya 1. Please try the notebook below. It provides two methods for generating a TensorFlow Lite (tflite) model. 2. I have tested it with the tiny model, and it works...
Hi @wompwomp13, sorry for late reply. Did you integrate it in your app or still facing issue.?
Magic number can be any four character string like "ABCD". It can be changed to your favourite string. But, need to check same string while loading vocab for validation.
Wave file should have below format 16K sampling rate, mono channel, 16bits bit depth. It can transcribe upto 30 seconds of audio data at a time.
For multilingual inputs, Whisper uses a custom tokenizer. For English only inputs, it uses the standard GPT-2 tokenizer which are both accessible through the open source [Whisper Python package](https://github.com/openai/whisper/blob/main/whisper/tokenizer.py#L361) Multilingual...
Do you mean transcription from one specific language to English?
I think larger models support this feature. Need to check