Create final model!

README.md

@@ -1,3 +1,145 @@
 ---
-license: mit
+language:
+- en
+thumbnail:
+tags:
+- audio-classification
+- speechbrain
+- embeddings
+- Accent
+- Identification
+- pytorch
+- ECAPA-TDNN
+- TDNN
+- CommonAccent
+license: "mit"
+datasets:
+- CommonVoice
+metrics:
+- Accuracy
+widget:
+- example_title: Australian English
+  src: australia_1.wav
+- example_title: African English
+  src: african_1.wav
+- example_title: Canadian English
+  src: canada_1.wav
 ---
+
+
+<iframe src="https://ghbtns.com/github-btn.html?user=speechbrain&repo=speechbrain&type=star&count=true&size=large&v=2" frameborder="0" scrolling="0" width="170" height="30" title="GitHub"></iframe>
+<br/><br/>
+
+# Accent Identification from Speech Recordings with ECAPA embeddings on CommonAccent
+
+This repository provides all the necessary tools to perform accent identification from speech recordings with SpeechBrain.
+The system uses a model pretrained on the CommonAccent dataset in English (16 accents).
+The provided system can recognize the following 16 languages from short speech recordings:
+
+```
+african australia bermuda canada england hongkong indian ireland malaysia newzealand philippines scotland singapore southatlandtic us wales
+```
+
+### To UPDATE ALL BELOW
+
+For a better experience, we encourage you to learn more about
+[SpeechBrain](https://speechbrain.github.io). The given model performance on the test set is:
+
+| Release | Accuracy (%)
+|:-------------:|:--------------:|
+| 30-06-21 | 85.0 | 
+
+
+## Pipeline description
+This system is composed of an ECAPA model coupled with statistical pooling. A classifier, trained with Categorical Cross-Entropy Loss, is applied on top of that.
+
+The system is trained with recordings sampled at 16kHz (single channel).
+The code will automatically normalize your audio (i.e., resampling + mono channel selection) when calling *classify_file* if needed. Make sure your input tensor is compliant with the expected sampling rate if you use *encode_batch* and *classify_batch*.
+
+## Install SpeechBrain
+
+First of all, please install SpeechBrain with the following command:
+
+```
+pip install speechbrain
+```
+
+Please notice that we encourage you to read our tutorials and learn more about
+[SpeechBrain](https://speechbrain.github.io).
+
+### Perform Language  Identification from Speech Recordings
+
+```python
+import torchaudio
+from speechbrain.pretrained import EncoderClassifier
+classifier = EncoderClassifier.from_hparams(source="speechbrain/lang-id-commonlanguage_ecapa", savedir="pretrained_models/lang-id-commonlanguage_ecapa")
+# Italian Example
+out_prob, score, index, text_lab = classifier.classify_file('speechbrain/lang-id-commonlanguage_ecapa/example-it.wav')
+print(text_lab)
+
+# French Example
+out_prob, score, index, text_lab = classifier.classify_file('speechbrain/lang-id-commonlanguage_ecapa/example-fr.wav')
+print(text_lab)
+```
+
+### Inference on GPU
+To perform inference on the GPU, add  `run_opts={"device":"cuda"}`  when calling the `from_hparams` method.
+
+### Training
+The model was trained with SpeechBrain (a02f860e).
+To train it from scratch follow these steps:
+1. Clone SpeechBrain:
+```bash
+git clone https://github.com/speechbrain/speechbrain/
+```
+2. Install it:
+```
+cd speechbrain
+pip install -r requirements.txt
+pip install -e .
+```
+
+3. Run Training:
+```
+cd recipes/CommonLanguage/lang_id
+python train.py hparams/train_ecapa_tdnn.yaml --data_folder=your_data_folder
+```
+
+You can find our training results (models, logs, etc) [here](https://drive.google.com/drive/folders/1sD2u0MhSmJlx_3RRgwsYzevX81RM8-WE?usp=sharing).
+
+### Limitations
+The SpeechBrain team does not provide any warranty on the performance achieved by this model when used on other datasets.
+
+#### Referencing ECAPA
+```@inproceedings{DBLP:conf/interspeech/DesplanquesTD20,
+  author    = {Brecht Desplanques and
+               Jenthe Thienpondt and
+               Kris Demuynck},
+  editor    = {Helen Meng and
+               Bo Xu and
+               Thomas Fang Zheng},
+  title     = {{ECAPA-TDNN:} Emphasized Channel Attention, Propagation and Aggregation
+               in {TDNN} Based Speaker Verification},
+  booktitle = {Interspeech 2020},
+  pages     = {3830--3834},
+  publisher = {{ISCA}},
+  year      = {2020},
+}
+```
+
+
+# **Citing SpeechBrain**
+Please, cite SpeechBrain if you use it for your research or business.
+
+
+```bibtex
+@misc{speechbrain,
+  title={{SpeechBrain}: A General-Purpose Speech Toolkit},
+  author={Mirco Ravanelli and Titouan Parcollet and Peter Plantinga and Aku Rouhe and Samuele Cornell and Loren Lugosch and Cem Subakan and Nauman Dawalatabad and Abdelwahab Heba and Jianyuan Zhong and Ju-Chieh Chou and Sung-Lin Yeh and Szu-Wei Fu and Chien-Feng Liao and Elena Rastorgueva and François Grondin and William Aris and Hwidong Na and Yan Gao and Renato De Mori and Yoshua Bengio},
+  year={2021},
+  eprint={2106.04624},
+  archivePrefix={arXiv},
+  primaryClass={eess.AS},
+  note={arXiv:2106.04624}
+}
+```

accent_encoder.txt

@@ -0,0 +1,18 @@
+'england' => 0
+'us' => 1
+'canada' => 2
+'australia' => 3
+'indian' => 4
+'scotland' => 5
+'ireland' => 6
+'african' => 7
+'malaysia' => 8
+'newzealand' => 9
+'southatlandtic' => 10
+'bermuda' => 11
+'philippines' => 12
+'hongkong' => 13
+'wales' => 14
+'singapore' => 15
+================
+'starting_index' => 0

classifier.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:146a2c6cb236e387b24972797ed9aebb3b54b09b33a072bee87eb3576bd88c01
+size 13172

config.json

@@ -0,0 +1,3 @@
+{
+  brain_interface": "EncoderClassifier"
+}

embedding_model.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ffa5ac9c0ec21fd6677fa8a39b3b045f182236f5c312a8edca385dfa79e7e3c
+size 83313275

hyperparams.yaml

@@ -0,0 +1,55 @@
+# ############################################################################
+# Model: ECAPA-TDNN for Accent Identification
+# ############################################################################
+
+# Pretrain folder (HuggingFace)
+pretrained_path: Jzuluaga/accent-id-commonaccent_ecapa
+
+# Feature parameters
+n_mels: 80
+
+# Output parameters
+n_languages: 16 # Possible languages in the dataset
+emb_dim: 192 # dimensionality of the embeddings
+
+# Model params
+compute_features: !new:speechbrain.lobes.features.Fbank
+    n_mels: !ref <n_mels>
+
+mean_var_norm: !new:speechbrain.processing.features.InputNormalization
+    norm_type: sentence
+    std_norm: False
+
+# Embedding Model
+embedding_model: !new:speechbrain.lobes.models.ECAPA_TDNN.ECAPA_TDNN
+    input_size: !ref <n_mels>
+    activation: !name:torch.nn.LeakyReLU
+    channels: [1024, 1024, 1024, 1024, 3072]
+    kernel_sizes: [5, 3, 3, 3, 1]
+    dilations: [1, 2, 3, 4, 1]
+    attention_channels: 128
+    lin_neurons: !ref <emb_dim>
+
+# Classifier based on cosine distance
+classifier: !new:speechbrain.lobes.models.ECAPA_TDNN.Classifier
+    input_size: !ref <emb_dim>
+    out_neurons: !ref <n_languages>
+
+modules:
+    compute_features: !ref <compute_features>
+    mean_var_norm: !ref <mean_var_norm>
+    embedding_model: !ref <embedding_model>
+    classifier: !ref <classifier>
+
+label_encoder: !new:speechbrain.dataio.encoder.CategoricalEncoder
+
+pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
+    loadables:
+        embedding_model: !ref <embedding_model>
+        classifier: !ref <classifier>
+        label_encoder: !ref <label_encoder>
+    paths:
+        embedding_model: !ref <pretrained_path>/embedding_model.ckpt
+        classifier: !ref <pretrained_path>/classifier.ckpt
+        label_encoder: !ref <pretrained_path>/accent_encoder.txt
+

normalizer_input.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f39c561f5e6e0030621263350a2c079cfcb675fb82b635402f57fdb7e1e21a5
+size 1163