mkenfenheuer/whisper-medium-cv11-german-ct2
<style> img { display: inline; } </style>
This model is a converted version of bofenghuang/whisper-medium-cv11-german converted to ctranslate2.
Fine-tuned whisper-medium model for ASR in German
This model is a fine-tuned version of openai/whisper-medium, trained on the mozilla-foundation/commonvoice11_0 de dataset. When using the model make sure that your speech input is also sampled at 16Khz. This model also predicts casing and punctuation.
Performance
Below are the WERs of the pre-trained models on the [Common Voice 9.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_9_0). These results are reported in the original [paper](https://cdn.openai.com/papers/whisper.pdf).
Below are the WERs of the fine-tuned models on the [Common Voice 11.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_11_0).
Usage
Inference with 🤗 Pipeline
import torch
from datasets import load_dataset
from transformers import pipeline
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# Load pipeline
pipe = pipeline("automatic-speech-recognition", model="bofenghuang/whisper-medium-cv11-german", device=device)
# NB: set forced_decoder_ids for generation utils
pipe.model.config.forced_decoder_ids = pipe.tokenizer.get_decoder_prompt_ids(language="de", task="transcribe")
# Load data
ds_mcv_test = load_dataset("mozilla-foundation/common_voice_11_0", "de", split="test", streaming=True)
test_segment = next(iter(ds_mcv_test))
waveform = test_segment["audio"]
# NB: decoding option
# limit the maximum number of generated tokens to 225
pipe.model.config.max_length = 225 + 1
# sampling
# pipe.model.config.do_sample = True
# beam search
# pipe.model.config.num_beams = 5
# return
# pipe.model.config.return_dict_in_generate = True
# pipe.model.config.output_scores = True
# pipe.model.config.num_return_sequences = 5
# Run
generated_sentences = pipe(waveform)["text"]Inference with 🤗 low-level APIs
import torch
import torchaudio
from datasets import load_dataset
from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# Load model
model = AutoModelForSpeechSeq2Seq.from_pretrained("bofenghuang/whisper-medium-cv11-german").to(device)
processor = AutoProcessor.from_pretrained("bofenghuang/whisper-medium-cv11-german", language="german", task="transcribe")
# NB: set forced_decoder_ids for generation utils
model.config.forced_decoder_ids = processor.get_decoder_prompt_ids(language="de", task="transcribe")
# 16_000
model_sample_rate = processor.feature_extractor.sampling_rate
# Load data
ds_mcv_test = load_dataset("mozilla-foundation/common_voice_11_0", "de", split="test", streaming=True)
test_segment = next(iter(ds_mcv_test))
waveform = torch.from_numpy(test_segment["audio"]["array"])
sample_rate = test_segment["audio"]["sampling_rate"]
# Resample
if sample_rate != model_sample_rate:
resampler = torchaudio.transforms.Resample(sample_rate, model_sample_rate)
waveform = resampler(waveform)
# Get feat
inputs = processor(waveform, sampling_rate=model_sample_rate, return_tensors="pt")
input_features = inputs.input_features
input_features = input_features.to(device)
# Generate
generated_ids = model.generate(inputs=input_features, max_new_tokens=225) # greedy
# generated_ids = model.generate(inputs=input_features, max_new_tokens=225, num_beams=5) # beam search
# Detokenize
generated_sentences = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
# Normalise predicted sentences if necessary