GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ZeeshanGeoPk
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
ZeeshanGeoPk
Model Tree
Input Modalities
Output Modalities
Supported Functionality
This repository contains a fine-tuned Whisper ASR (Automatic Speech Recognition) model for the Haitian language. The model is hosted on Hugging Face and is ready for use.
The model achieved a Word Error Rate (WER) of 0.19126, indicating high accuracy in transcribing spoken Haitian to written text.
The model was trained with a learning rate of 1e-5.
You can use this model directly from the Hugging Face Model Hub. Here's a simple example in Python:
text
from transformers import WhisperProcessor, WhisperForConditionalGenerationimport torchaudio# load model and processorprocessor = WhisperProcessor.from_pretrained("ZeeshanGeoPk/haitian-speech-to-text")model = WhisperForConditionalGeneration.from_pretrained("ZeeshanGeoPk/haitian-speech-to-text")# read audio filessample_path = "path/to/audio.wav"# load audio file using torchaudiowaveform, sample_rate = torchaudio.load(sample_path)# resample if needed (Whisper model requires 16kHz)if sample_rate != 16000: resampler = torchaudio.transforms.Resample(sample_rate, 16000) waveform = resampler(waveform) sample_rate = 16000# ensure mono channelif waveform.shape[0] > 1: waveform = waveform.mean(dim=0, keepdim=True)# process audio using Whisper processorinput_features = processor(waveform.numpy(), sampling_rate=sample_rate, return_tensors="pt").input_features# generate token idspredicted_ids = model.generate(input_features)# decode token ids to texttranscription = processor.batch_decode(predicted_ids, skip_special_tokens=True)print(transcription)