Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,62 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
+
language:
|
4 |
+
- de
|
5 |
+
library_name: transformers
|
6 |
+
pipeline_tag: automatic-speech-recognition
|
7 |
---
|
8 |
+
|
9 |
+
# whisper-tiny-german
|
10 |
+
|
11 |
+
This model is a German Speech Recognition model based on the [whisper-tiny](https://huggingface.co/openai/whisper-tiny) model.
|
12 |
+
The model weights count 756M parameters and with a size of 1.51GB in bfloat16 format.
|
13 |
+
|
14 |
+
As a follow-up to the [Whisper large v3 german](https://huggingface.co/primeline/whisper-large-v3-german) we decided to create a distilled version for a faster inference with minimal quality loss.
|
15 |
+
|
16 |
+
## Intended uses & limitations
|
17 |
+
|
18 |
+
The model is intended to be used for German speech recognition tasks.
|
19 |
+
It can be used as local transkription service or as a part of a larger pipeline for speech recognition tasks.
|
20 |
+
While counting only half of the parameters of the large model, the quality is still very good and can be used for most tasks.
|
21 |
+
The latency is low enough to be used in real-time applications when using optimization toolkits like tensorrt.
|
22 |
+
|
23 |
+
## Dataset
|
24 |
+
|
25 |
+
The dataset used for training is a filtered subset of the [Common Voice](https://huggingface.co/datasets/common_voice) dataset, multilingual librispeech and some internal data.
|
26 |
+
The data was filtered and double checked for quality and correctness.
|
27 |
+
We did some normalization to the text data, especially for casing and punctuation.
|
28 |
+
|
29 |
+
|
30 |
+
## Model family
|
31 |
+
|
32 |
+
| Model | Parameters | link |
|
33 |
+
|----------------------------------|------------|--------------------------------------------------------------|
|
34 |
+
| Whisper large v3 german | 1.54B | [link](https://huggingface.co/primeline/whisper-large-v3-german) |
|
35 |
+
| Distil-whisper large v3 german | 756M | [link](https://huggingface.co/primeline/distil-whisper-large-v3-german) |
|
36 |
+
| tiny whisper | 37.8M | [link](https://huggingface.co/primeline/whisper-tiny-german) |
|
37 |
+
|
38 |
+
### Training hyperparameters
|
39 |
+
|
40 |
+
The following hyperparameters were used during training:
|
41 |
+
- learning_rate: 3e-05
|
42 |
+
- total_train_batch_size: 512
|
43 |
+
- num_epochs: 5.0
|
44 |
+
|
45 |
+
### Framework versions
|
46 |
+
|
47 |
+
- Transformers 4.39.3
|
48 |
+
- Pytorch 2.3.0a0+ebedce2
|
49 |
+
- Datasets 2.18.0
|
50 |
+
- Tokenizers 0.15.2
|
51 |
+
|
52 |
+
## [About us](https://primeline-ai.com/en/)
|
53 |
+
|
54 |
+
[![primeline AI](https://primeline-ai.com/wp-content/uploads/2024/02/pl_ai_bildwortmarke_original.svg)](https://primeline-ai.com/en/)
|
55 |
+
|
56 |
+
|
57 |
+
Your partner for AI infrastructure in Germany <br>
|
58 |
+
Experience the powerful AI infrastructure that drives your ambitions in Deep Learning, Machine Learning & High-Performance Computing. Optimized for AI training and inference.
|
59 |
+
|
60 |
+
|
61 |
+
|
62 |
+
Model author: [Florian Zimmermeister](https://huggingface.co/flozi00)
|