Upload 3 files

README.md

@@ -0,0 +1,132 @@
+---
+language:
+- fa
+thumbnail: null
+pipeline_tag: automatic-speech-recognition
+tags:
+- whisper
+- pytorch
+- speechbrain
+- Transformer
+- hf-asr-leaderboard
+license: apache-2.0
+datasets:
+- commonvoice
+metrics:
+- wer
+- cer
+model-index:
+- name: asr-whisper-large-v2-commonvoice-ar
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: CommonVoice 10.0 (Persian)
+      type: mozilla-foundation/common_voice_10_0
+      config: fa
+      split: test
+      args:
+        language: fa
+    metrics:
+    - name: Test WER
+      type: wer
+      value: '31.75'
+---
+
+<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/>
+
+# whisper large-v2 fine-tuned on CommonVoice Persian
+
+This repository provides all the necessary tools to perform automatic speech
+recognition from an end-to-end whisper model fine-tuned on CommonVoice (Persian Language) within
+SpeechBrain. For a better experience, we encourage you to learn more about
+[SpeechBrain](https://speechbrain.github.io).
+
+The performance of the model is the following:
+
+| Release | Test CER | Test WER | GPUs |
+|:-------------:|:--------------:|:--------------:| :--------:|
+| 01-02-23 | 9.38 |  31.75 | 1xV100 16GB |
+
+## Pipeline description
+
+This ASR system is composed of whisper encoder-decoder blocks:
+- The pretrained whisper-large-v2 encoder is frozen.
+- The pretrained Whisper tokenizer is used.
+- A pretrained Whisper-large-v2 decoder ([openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2)) is finetuned on CommonVoice Fa.
+The obtained final acoustic representation is given to the greedy decoder. 
+
+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 *transcribe_file* if needed.
+
+## Install SpeechBrain
+
+First of all, please install tranformers and SpeechBrain with the following command:
+
+```
+pip install speechbrain transformers
+```
+
+Please notice that we encourage you to read our tutorials and learn more about
+[SpeechBrain](https://speechbrain.github.io).
+
+### Transcribing your own audio files (in Persian)
+
+```python
+
+from speechbrain.pretrained import WhisperASR
+
+asr_model = WhisperASR.from_hparams(source="speechbrain/asr-whisper-large-v2-commonvoice-fa", savedir="retrained_models/asr-whisper-large-v2-commonvoice-fa")
+asr_model.transcribe_file("speechbrain/asr-whisper-large-v2-commonvoice-ar/example-fa.mp3")
+
+
+```
+### 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.
+To train it from scratch follow these steps:
+1. Clone SpeechBrain:
+```bash
+git clone https://github.com/speechbrain/speechbrain/
+```
+2. Install it:
+```bash
+cd speechbrain
+pip install -r requirements.txt
+pip install -e .
+```
+
+3. Run Training:
+```bash
+cd recipes/CommonVoice/ASR/transformer/
+python train_with_whisper.py hparams/train_fa_hf_whisper.yaml --data_folder=your_data_folder
+```
+
+You can find our training results (models, logs, etc) [here](https://drive.google.com/drive/folders/1nzMMYmB5SxMKsFUk-rM9_ijcqzia8pX7).
+
+### Limitations
+The SpeechBrain team does not provide any warranty on the performance achieved by this model when used on other datasets.
+
+#### Referencing SpeechBrain
+
+```
+@misc{SB2021,
+    author = {Ravanelli, Mirco and Parcollet, Titouan and Rouhe, Aku and Plantinga, Peter and Rastorgueva, Elena and Lugosch, Loren and Dawalatabad, Nauman and Ju-Chieh, Chou and Heba, Abdel and Grondin, Francois and Aris, William and Liao, Chien-Feng and Cornell, Samuele and Yeh, Sung-Lin and Na, Hwidong and Gao, Yan and Fu, Szu-Wei and Subakan, Cem and De Mori, Renato and Bengio, Yoshua },
+    title = {SpeechBrain},
+    year = {2021},
+    publisher = {GitHub},
+    journal = {GitHub repository},
+    howpublished = {\\\\url{https://github.com/speechbrain/speechbrain}},
+  }
+```
+
+#### About SpeechBrain
+SpeechBrain is an open-source and all-in-one speech toolkit. It is designed to be simple, extremely flexible, and user-friendly. Competitive or state-of-the-art performance is obtained in various domains.
+
+Website: https://speechbrain.github.io/
+
+GitHub: https://github.com/speechbrain/speechbrain

config.json

@@ -0,0 +1,3 @@
+{
+  "speechbrain_interface": "WhisperASR"
+}

hyperparams.yaml

@@ -0,0 +1,77 @@
+# ################################
+# Model: Whisper (Encoder-Decoder) + NLL
+# Augmentation: TimeDomainSpecAugment
+# Authors: Pooneh Mousavi 2022
+# ################################
+
+
+# URL for the biggest Fairseq english whisper model.
+whisper_hub: openai/whisper-large-v2
+
+# Normalize inputs with
+# the same normalization done in the paper. Refer to Appendix C for further information.
+normalized_transcripts: True
+
+
+language: persian 
+
+auto_mix_prec: False
+sample_rate: 16000
+
+# These values are only used for the searchers.
+# They needs to be hardcoded and should not be changed with Whisper.
+# They are used as part of the searching process.
+# The bos token of the searcher will be timestamp_index
+# and will be concatenated with the bos, language and task tokens.
+timestamp_index: 50363
+eos_index: 50257
+bos_index: 50258
+
+# Decoding parameters
+min_decode_ratio: 0.0
+max_decode_ratio: 0.1
+test_beam_size: 8
+
+# Model parameters
+freeze_whisper: True
+freeze_encoder: True
+
+
+
+whisper: !new:speechbrain.lobes.models.huggingface_whisper.HuggingFaceWhisper
+    source: !ref <whisper_hub>
+    freeze: !ref <freeze_whisper>
+    freeze_encoder: !ref <freeze_encoder>
+    save_path: whisper_checkpoints
+    encoder_only:  False
+
+
+
+decoder: !new:speechbrain.decoders.seq2seq.S2SWhisperGreedySearch
+    model: !ref <whisper>
+    bos_index: !ref <timestamp_index>
+    eos_index: !ref <eos_index>
+    min_decode_ratio: !ref <min_decode_ratio>
+    max_decode_ratio: !ref <max_decode_ratio>
+
+# test_beam_searcher: !new:speechbrain.decoders.seq2seq.S2SWhisperBeamSearch
+#     module: [!ref <whisper>]
+#     bos_index: !ref <timestamp_index>
+#     eos_index: !ref <eos_index>
+#     min_decode_ratio: !ref <min_decode_ratio>
+#     max_decode_ratio: !ref <max_decode_ratio>
+#     beam_size: !ref <test_beam_size>
+
+
+
+
+
+modules:
+    whisper: !ref <whisper>
+    decoder:  !ref <decoder>
+ 
+
+pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
+    loadables:
+        whisper: !ref <whisper>
+