[S2T] Whisper transcription cannot choose languages #2827
Description
For support and discussions, please use our Discourse forums.
If you've found a bug then please create an issue with the following information:
Describe the bug
Even if users specify a language, program will still detect language by itself.
To Reproduce
import paddle
from paddlespeech.cli.whisper import WhisperExecutor
audio_file = 'path/to/test.wav'
whisper_executor = WhisperExecutor()
result = whisper_executor(
model='whisper',
task='transcribe',
size='tiny',
sample_rate=16000,
config=None, # Set `config` and `ckpt_path` to None to use pretrained model.
ckpt_path=None,
audio_file=audio_file,
language='fr',
device=paddle.get_device())Screenshots