Updated model information
Browse files
README.md
CHANGED
@@ -1,3 +1,40 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
+
datasets:
|
4 |
+
- mozilla-foundation/common_voice_10_0
|
5 |
+
base_model:
|
6 |
+
- facebook/wav2vec2-xls-r-300m
|
7 |
+
tags:
|
8 |
+
- pytorch
|
9 |
+
- phoneme-recognition
|
10 |
+
pipeline_tag: automatic-speech-recognition
|
11 |
---
|
12 |
+
|
13 |
+
Model Information
|
14 |
+
=================
|
15 |
+
|
16 |
+
Allophant is a multilingual phoneme recognizer trained on spoken sentences in 34 languages, capable of generalizing zero-shot to unseen phoneme inventories.
|
17 |
+
|
18 |
+
The model is based on [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) and was pre-trained on a subset of the [Common Voice Corpus 10.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_10_0) transcribed with [eSpeak NG](https://github.com/espeak-ng/espeak-ng).
|
19 |
+
|
20 |
+
| Model Name | UCLA Phonetic Corpus (PER) | UCLA Phonetic Corpus (AER) | Common Voice (PER) | Common Voice (AER) |
|
21 |
+
| ---------------- | ---------: | ---------: | -------: | -------: |
|
22 |
+
| [Multitask](https://huggingface.co/kgnlp/allophant) | **45.62%** | 19.44% | **34.34%** | **8.36%** |
|
23 |
+
| **Hierarchical** | 46.09% | **19.18%** | 34.35% | 8.56% |
|
24 |
+
| [Multitask Shared](https://huggingface.co/kgnlp/allophant-shared) | 46.05% | 19.52% | 41.20% | 8.88% |
|
25 |
+
| [Baseline Shared](https://huggingface.co/kgnlp/allophant-baseline-shared) | 48.25% | - | 45.35% | - |
|
26 |
+
| [Baseline](https://huggingface.co/kgnlp/allophant-baseline) | 57.01% | - | 46.95% | - |
|
27 |
+
|
28 |
+
Note that our baseline models were trained without phonetic feature classifiers and therefore only support phoneme recognition.
|
29 |
+
|
30 |
+
Citation
|
31 |
+
========
|
32 |
+
|
33 |
+
```bibtex
|
34 |
+
@inproceedings{glocker2023allophant,
|
35 |
+
title={Allophant: Cross-lingual Phoneme Recognition with Articulatory Attributes},
|
36 |
+
author={Glocker, Kevin and Herygers, Aaricia and Georges, Munir},
|
37 |
+
year={2023},
|
38 |
+
booktitle={{Proc. Interspeech 2023}},
|
39 |
+
month={8}}
|
40 |
+
```
|