Update README.md (#2)

- Update README.md (a07e496f146647196fe42a5bd59785afa69b1867)


Co-authored-by: He Huang <steveheh@users.noreply.huggingface.co>

README.md

@@ -23,9 +23,10 @@ datasets:
 thumbnail: null
 tags:
 - automatic-speech-recognition
+- automatic-speech-translation
 - speech
 - audio
-- Transducer
+- Transformer
 - FastConformer
 - Conformer
 - pytorch
@@ -37,62 +38,8 @@ widget:
 - example_title: Librispeech sample 2
   src: https://cdn-media.huggingface.co/speech_samples/sample2.flac
 model-index:
-- name: parakeet_rnnt_1.1b
+- name: canary-1b
   results:
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: AMI (Meetings test)
-      type: edinburghcstr/ami
-      config: ihm
-      split: test
-      args:
-        language: en
-    metrics:
-    - name: Test WER
-      type: wer
-      value: 17.10
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: Earnings-22
-      type: revdotcom/earnings22
-      split: test
-      args:
-        language: en
-    metrics:
-    - name: Test WER
-      type: wer
-      value: 14.11
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: GigaSpeech
-      type: speechcolab/gigaspeech
-      split: test
-      args:
-        language: en
-    metrics:
-    - name: Test WER
-      type: wer
-      value: 9.96
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: LibriSpeech (clean)
-      type: librispeech_asr
-      config: other
-      split: test
-      args:
-        language: en
-    metrics:
-    - name: Test WER
-      type: wer
-      value: 1.46
   - task:
       name: Automatic Speech Recognition
       type: automatic-speech-recognition
@@ -106,7 +53,7 @@ model-index:
     metrics:
     - name: Test WER
       type: wer
-      value: 2.47
+      value: 2.89
   - task:
       type: Automatic Speech Recognition
       name: automatic-speech-recognition
@@ -120,27 +67,13 @@ model-index:
     metrics:
     - name: Test WER
       type: wer
-      value: 3.11
+      value: 4.79
   - task:
       type: Automatic Speech Recognition
       name: automatic-speech-recognition
     dataset:
-      name: tedlium-v3
-      type: LIUM/tedlium
-      config: release1
-      split: test
-      args:
-        language: en
-    metrics:
-    - name: Test WER
-      type: wer
-      value: 3.92
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: Vox Populi
-      type: facebook/voxpopuli
+      name: Mozilla Common Voice 16.1
+      type: mozilla-foundation/common_voice_16_1
       config: en
       split: test
       args:
@@ -148,24 +81,10 @@ model-index:
     metrics:
     - name: Test WER
       type: wer
-      value: 5.39
-  - task:
-      type: Automatic Speech Recognition
-      name: automatic-speech-recognition
-    dataset:
-      name: Mozilla Common Voice 9.0
-      type: mozilla-foundation/common_voice_9_0
-      config: en
-      split: test
-      args:
-        language: en
-    metrics:
-    - name: Test WER
-      type: wer
-      value: 5.79
-  
+      value: 3.99
 metrics:
 - wer
+- bleu
 pipeline_tag: automatic-speech-recognition
 ---
 
@@ -185,94 +104,172 @@ img {
 | [![Language](https://img.shields.io/badge/Language-es-lightgrey#model-badge)](#datasets)
 | [![Language](https://img.shields.io/badge/Language-fr-lightgrey#model-badge)](#datasets)
 
-CHANGE FROM HERE
-`parakeet-rnnt-1.1b` is an ASR model that transcribes speech in lower case English alphabet. This model is jointly developed by [NVIDIA NeMo](https://github.com/NVIDIA/NeMo) and [Suno.ai](https://www.suno.ai/) teams.
-It is an XXL version of FastConformer Transducer [1] (around 1.1B parameters) model.
-See the [model architecture](#model-architecture) section and [NeMo documentation](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/asr/models.html#fast-conformer) for complete architecture details.
+NVIDIA NeMo Canary is a family of multi-lingual multi-tasking models that achieves state-of-the art performance on multiple benchmarks. With 1 billion parameters, Canary-1B supports automatic speech-to-text recognition (ASR) in 4 languages (English, German, French, Spanish) and translation from English to German/French/Spanish and from German/French/Spanish to English with or without punctuation and capitalization (PnC). 
+
+## Model Architecture
+Canary is an encoder-decoder model with FastConformer [1] encoder and Transformer Decoder [2]. 
+With audio features extracted from the encoder, task tokens such as `<source language>`, `<target language>`, `<task>` and `<toggle PnC>` 
+are fed into the Transformer Decoder to trigger the text generation process. Canary uses a concatenated tokenizer from individual 
+SentencePiece [3] tokenizers of each language, which makes it easy to scale up to more languages. 
+The Canay-1B model has 24 encoder layers and 24 layers of decoder layers in total.
+
+
 
-## NVIDIA NeMo: Training
+## NVIDIA NeMo
 
 To train, fine-tune or play with the model you will need to install [NVIDIA NeMo](https://github.com/NVIDIA/NeMo). We recommend you install it after you've installed latest PyTorch version.
 ```
 pip install nemo_toolkit['all']
-``` 
+```
+
 
 ## How to Use this Model
 
-The model is available for use in the NeMo toolkit [3], and can be used as a pre-trained checkpoint for inference or for fine-tuning on another dataset.
+The model is available for use in the NeMo toolkit [4], and can be used as a pre-trained checkpoint for inference or for fine-tuning on another dataset.
 
-### Automatically instantiate the model
+### Loading the Model
 
 ```python
-import nemo.collections.asr as nemo_asr
-asr_model = nemo_asr.models.EncDecRNNTBPEModel.from_pretrained(model_name="nvidia/parakeet-rnnt-1.1b")
+from nemo.collections.asr.models import EncDecMultiTaskModel
+
+# load model
+canary_model = EncDecMultiTaskModel.from_pretrained('nvidia/canary-1b')
+
+# update dcode params
+decode_cfg = canary_model.cfg.decoding
+decode_cfg.beam.beam_size = 5  # default is greedy with beam_size=1
+canary_model.change_decoding_strategy(decode_cfg)
 ```
 
-### Transcribing using Python
-First, let's get a sample
+### Input Format
+The input to the model can be a directory containing audio files, in which case the model will perform ASR on English and produces text with punctuation and capitalization:
+
+```python
+predicted_text = canary_model.trancribe(
+    audio_dir="<path to directory containing audios>",
+    batch_size=16,  # batch size to run the inference with
+)
 ```
-wget https://dldata-public.s3.us-east-2.amazonaws.com/2086-149220-0033.wav
+
+or use:
+
+```bash
+python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py 
+ pretrained_name="nvidia/canary-1b" 
+ audio_dir="<path to audio directory>" 
 ```
-Then simply do:
+
+Another recommended option is to use a json manifest as input, where each line in the file is a dictionary containing the following fields:
+```yaml
+# Example of a line in input_manifest.json
+{
+    "audio_filepath": "/path/to/audio.wav",  # path to the audio file
+    "duration": 10000.0,  # duration of the audio
+    "taskname": "asr",  # use "s2t_translation" for AST
+    "source_lang": "en",  # Set `source_lang`=`target_lang` for ASR, choices=['en','de','es','fr']
+    "target_lang": "de",  # choices=['en','de','es','fr']
+    "pnc": yes,  # whether to have PnC output, choices=['yes', 'no'] 
+}
 ```
-asr_model.transcribe(['2086-149220-0033.wav'])
+
+and then use:
+```python
+predicted_text = canary_model.trancribe(
+    paths2audio_files="<path to input manifest file>",
+    batch_size=16,  # batch size to run the inference with
+)
 ```
 
-### Transcribing many audio files
+or use:
 
-```shell
+```bash
 python [NEMO_GIT_FOLDER]/examples/asr/transcribe_speech.py 
- pretrained_name="nvidia/parakeet-rnnt-1.1b" 
- audio_dir="<DIRECTORY CONTAINING AUDIO FILES>"
+ pretrained_name="nvidia/canary-1b" 
+ dataset_manifest="<path to manifest file>" 
+```
+
+### Automatic Speech-to-text Recognition (ASR)
+
+An example manifest for transcribing English audios can be:
+
+```yaml
+# Example of a line in input_manifest.json
+{
+    "audio_filepath": "/path/to/audio.wav",  # path to the audio file
+    "duration": 10000.0,  # duration of the audio
+    "taskname": "asr",  
+    "source_lang": "en", 
+    "target_lang": "en", 
+    "pnc": yes,  # whether to have PnC output, choices=['yes', 'no'] 
+}
+```
+
+
+### Automatic Speech-to-text Translation (AST)
+
+An example manifest for transcribing English audios into German text can be:
+
+```yaml
+# Example of a line in input_manifest.json
+{
+    "audio_filepath": "/path/to/audio.wav",  # path to the audio file
+    "duration": 10000.0,  # duration of the audio
+    "taskname": "s2t_translation",  
+    "source_lang": "en",  
+    "target_lang": "de",
+    "pnc": yes,  # whether to have PnC output, choices=['yes', 'no'] 
+}
 ```
 
+
 ### Input
 
-This model accepts 16000 Hz mono-channel audio (wav files) as input.
+This model accepts single channel (mono) audio sampled at 16000 Hz, along with the task/languages/PnC tags as input.
 
 ### Output
 
-This model provides transcribed speech as a string for a given audio sample.
+The model outputs the transcribed/translated text corresponding to the input audio, in the specified target language and with or without punctuation and capitalization.
 
-## Model Architecture
 
-FastConformer [1] is an optimized version of the Conformer model with 8x depthwise-separable convolutional downsampling. The model is trained in a multitask setup with a Transducer decoder (RNNT) loss. You may find more information on the details of FastConformer here: [Fast-Conformer Model](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/asr/models.html#fast-conformer).
 
 ## Training
 
-The NeMo toolkit [3] was used for training the models for over several hundred epochs. These model are trained with this [example script](https://github.com/NVIDIA/NeMo/blob/main/examples/asr/asr_transducer/speech_to_text_rnnt_bpe.py) and this [base config](https://github.com/NVIDIA/NeMo/blob/main/examples/asr/conf/fastconformer/fast-conformer_transducer_bpe.yaml).
+Canary-1B is trained using the  NVIDIA NeMo toolkit [4] for 150k steps with dynamic bucketing and a batch duration of 360s per GPU on 128 NVIDIA A100 80GB GPUs in 24 hrs. The model can be trained using this example script and base config.
 
 The tokenizers for these models were built using the text transcripts of the train set with this [script](https://github.com/NVIDIA/NeMo/blob/main/scripts/tokenizers/process_asr_text_tokenizer.py).
 
+
 ### Datasets
 
-The model was trained on 64K hours of English speech collected and prepared by NVIDIA NeMo and Suno teams.
+The Canary-1B model is trained on 70K hours of speech audio with transcriptions in their original languages for ASR, and machine-generated translations for each supported language for speech translation. 
 
-The training dataset consists of private subset with 40K hours of English speech plus 24K hours from the following public datasets:
+The training data contains 43K hours of English speech collected and prepared by NVIDIA NeMo and [Suno](https://suno.ai/) teams, and an inhouse subset with 27K hours of English/German/Spanish/French speech.
 
-- Librispeech 960 hours of English speech
-- Fisher Corpus
-- Switchboard-1 Dataset
-- WSJ-0 and WSJ-1
-- National Speech Corpus (Part 1, Part 6)
-- VCTK
-- VoxPopuli (EN)
-- Europarl-ASR (EN)
-- Multilingual Librispeech (MLS EN) - 2,000 hour subset
-- Mozilla Common Voice (v7.0)
-- People's Speech  - 12,000 hour subset
 
 ## Performance
 
-The performance of Automatic Speech Recognition models is measuring using Word Error Rate. Since this dataset is trained on multiple domains and a much larger corpus, it will generally perform better at transcribing audio in general.
+The ASR performance is measured with word error rate (WER) on different datasets, whereas the AST performance is measured with BLEU score. Predictions were generated using beam search with width 5 and length penalty 1.0.
+
+### ASR Performance (w/o PnC) 
+
+We use [MCV-16.1](https://commonvoice.mozilla.org/en/datasets) test sets on four languages, and process the groundtruth and predicted text with [whisper-normalizer](https://pypi.org/project/whisper-normalizer/).
+
 
-The following tables summarizes the performance of the available models in this collection with the Transducer decoder. Performances of the ASR models are reported in terms of Word Error Rate (WER%) with greedy decoding. 
+| **Version** | **Model**     | **En**   | **De**   | **Es**   | **Fr**   |
+|:---------:|:-----------:|:------:|:------:|:------:|:------:|
+| 1.23.0  | canary-1b | 7.97 | 4.61 | 3.99 | 6.53 |
 
-|**Version**|**Tokenizer**|**Vocabulary Size**|**AMI**|**Earnings-22**|**Giga Speech**|**LS test-clean**|**SPGI Speech**|**TEDLIUM-v3**|**Vox Populi**|**Common Voice**|
-|---------|-----------------------|-----------------|---------------|---------------|------------|-----------|-----|-------|------|------|
-| 1.22.0  | SentencePiece Unigram | 1024 | 17.10 | 14.11 | 9.96 | 1.46 | 2.47 | 3.11 | 3.92 | 5.39 | 5.79 |
 
-These are greedy WER numbers without external LM. More details on evaluation can be found at [HuggingFace ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard)
+More details on evaluation can be found at [HuggingFace ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard)
+
+### AST Performance
+
+We evaluate on the FLEURS test sets and use the native annotations with punctuation and capitalization.
+
+| **Version** | **Model** | **En->De** | **En->Es** | **En->Fr** | **De->En** | **Es->En** | **Fr->En** |
+|:-----------:|:---------:|:----------:|:----------:|:----------:|:----------:|:----------:|:----------:|
+| 1.23.0      | canary-1b | 22.66	      | 41.11      | 40.76      | 32.64      | 32.15      | 23.57      |
+
 
 ## NVIDIA Riva: Deployment
 
@@ -283,21 +280,20 @@ Additionally, Riva provides:
 * Best in class accuracy with run-time word boosting (e.g., brand and product names) and customization of acoustic model, language model, and inverse text normalization 
 * Streaming speech recognition, Kubernetes compatible scaling, and enterprise-grade support 
 
-Although this model isn’t supported yet by Riva, the [list of supported models is here](https://huggingface.co/models?other=Riva).  
+Although this model isn’t supported yet by Riva, the [list of supported models](https://huggingface.co/models?other=Riva) is here.  
 Check out [Riva live demo](https://developer.nvidia.com/riva#demos). 
 
+
 ## References
 [1] [Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition](https://arxiv.org/abs/2305.05084)
 
-[2] [Google Sentencepiece Tokenizer](https://github.com/google/sentencepiece)
-
-[3] [NVIDIA NeMo Toolkit](https://github.com/NVIDIA/NeMo)
+[2] [Attention is all you need](https://arxiv.org/abs/1706.03762)
 
-[4] [Suno.ai](https://suno.ai/)
+[3] [Google Sentencepiece Tokenizer](https://github.com/google/sentencepiece)
 
-[5] [HuggingFace ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard)
+[4] [NVIDIA NeMo Toolkit](https://github.com/NVIDIA/NeMo)
 
 
 ## Licence
 
-License to use this model is covered by the [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/). By downloading the public and release version of the model, you accept the terms and conditions of the [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/) license.
+License to use this model is covered by the [CC-BY-NC-4.0](https://creativecommons.org/licenses/by-nc/4.0/deed.en#:~:text=NonCommercial%20%E2%80%94%20You%20may%20not%20use,doing%20anything%20the%20license%20permits.). By downloading the public and release version of the model, you accept the terms and conditions of the CC-BY-NC-4.0 license.