Update README.md

README.md

@@ -147,318 +147,197 @@ metrics:
 - wer
 ---
 
-# **Parakeet TDT 0.6B V2 (En)**
+# Model Card for Model ID
 
-<style>
-img {
- display: inline;
-}
-</style>
+<!-- Provide a quick summary of what the model is/does. -->
 
-[![Model architecture](https://img.shields.io/badge/Model_Arch-FastConformer--TDT-blue#model-badge)](#model-architecture)
-| [![Model size](https://img.shields.io/badge/Params-0.6B-green#model-badge)](#model-architecture)
-| [![Language](https://img.shields.io/badge/Language-en-orange#model-badge)](#datasets)
+This modelcard aims to be a base template for new models. It has been generated using [this raw template](https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/templates/modelcard_template.md?plain=1).
 
+## Model Details
 
-## <span style="color:#466f00;">Description:</span>
+### Model Description
 
-`parakeet-tdt-0.6b-v2` is a 600-million-parameter automatic speech recognition (ASR) model designed for high-quality English transcription, featuring support for punctuation, capitalization, and accurate timestamp prediction. Try Demo here: https://huggingface.co/spaces/nvidia/parakeet-tdt-0.6b-v2 
+<!-- Provide a longer summary of what this model is. -->
 
-This XL variant of the FastConformer [1] architecture integrates the TDT [2] decoder and is trained with full attention, enabling efficient transcription of audio segments up to 24 minutes in a single pass. The model achieves an RTFx of 3380 on the HF-Open-ASR leaderboard with a batch size of 128. Note: *RTFx Performance may vary depending on dataset audio duration and batch size.*  
 
-**Key Features**
-- Accurate word-level timestamp predictions  
-- Automatic punctuation and capitalization  
-- Robust performance on spoken numbers, and song lyrics transcription 
 
-For more information, refer to the [Model Architecture](#model-architecture) section and the [NeMo documentation](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/asr/models.html#fast-conformer).
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
 
-This model is ready for commercial/non-commercial use.
+### Model Sources [optional]
 
+<!-- Provide the basic links for the model. -->
 
-## <span style="color:#466f00;">License/Terms of Use:</span>
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
 
-GOVERNING TERMS: Use of this model is governed by the [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/legalcode.en) license.
+## Uses
 
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
 
-### <span style="color:#466f00;">Deployment Geography:</span>
-Global
+### Direct Use
 
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
 
-### <span style="color:#466f00;">Use Case:</span>
+[More Information Needed]
 
-This model serves developers, researchers, academics, and industries building applications that require speech-to-text capabilities, including but not limited to: conversational AI, voice assistants, transcription services, subtitle generation, and voice analytics platforms.
+### Downstream Use [optional]
 
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
 
-### <span style="color:#466f00;">Release Date:</span>
+[More Information Needed]
 
-05/01/2025
+### Out-of-Scope Use
 
-### <span style="color:#466f00;">Model Architecture:</span>
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
 
-**Architecture Type**: 
+[More Information Needed]
 
-FastConformer-TDT
+## Bias, Risks, and Limitations
 
-**Network Architecture**:
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
 
-* This model was developed based on [FastConformer encoder](https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en/main/asr/models.html#fast-conformer) architecture[1] and TDT decoder[2]
-* This model has 600 million model parameters.
+[More Information Needed]
 
-### <span style="color:#466f00;">Input:</span>
-- **Input Type(s):** 16kHz Audio
-- **Input Format(s):** `.wav` and `.flac` audio formats
-- **Input Parameters:** 1D (audio signal)
-- **Other Properties Related to Input:**  Monochannel audio
+### Recommendations
 
-### <span style="color:#466f00;">Output:</span>
-- **Output Type(s):**  Text
-- **Output Format:**  String
-- **Output Parameters:**  1D (text)
-- **Other Properties Related to Output:** Punctuations and Capitalizations included.
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
 
-Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA's hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions. 
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
 
-## <span style="color:#466f00;">How to Use this Model:</span>
+## How to Get Started with the Model
 
-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.
-```bash
-pip install -U nemo_toolkit['asr']
-``` 
-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.
+Use the code below to get started with the model.
 
-#### Automatically instantiate the model
+[More Information Needed]
 
-```python
-import nemo.collections.asr as nemo_asr
-asr_model = nemo_asr.models.ASRModel.from_pretrained(model_name="nvidia/parakeet-tdt-0.6b-v2")
-```
+## Training Details
 
-#### Transcribing using Python
-First, let's get a sample
-```bash
-wget https://dldata-public.s3.us-east-2.amazonaws.com/2086-149220-0033.wav
-```
-Then simply do:
-```python
-output = asr_model.transcribe(['2086-149220-0033.wav'])
-print(output[0].text)
-```
+### Training Data
 
-#### Transcribing with timestamps
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
 
-To transcribe with timestamps:
-```python
-output = asr_model.transcribe(['2086-149220-0033.wav'], timestamps=True)
-# by default, timestamps are enabled for char, word and segment level
-word_timestamps = output[0].timestamp['word'] # word level timestamps for first sample
-segment_timestamps = output[0].timestamp['segment'] # segment level timestamps
-char_timestamps = output[0].timestamp['char'] # char level timestamps
+[More Information Needed]
 
-for stamp in segment_timestamps:
-    print(f"{stamp['start']}s - {stamp['end']}s : {stamp['segment']}")
-```
+### Training Procedure
 
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
 
-## <span style="color:#466f00;">Software Integration:</span>
+#### Preprocessing [optional]
 
-**Runtime Engine(s):**
-* NeMo 2.2  
+[More Information Needed]
 
 
-**Supported Hardware Microarchitecture Compatibility:** 
-* NVIDIA Ampere
-* NVIDIA Blackwell  
-* NVIDIA Hopper
-* NVIDIA Volta
+#### Training Hyperparameters
 
-**[Preferred/Supported] Operating System(s):**
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
 
-- Linux
+#### Speeds, Sizes, Times [optional]
 
-**Hardware Specific Requirements:**
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
 
-Atleast 2GB RAM for model to load. The bigger the RAM, the larger audio input it supports.
+[More Information Needed]
 
-#### Model Version
+## Evaluation
 
-Current version: parakeet-tdt-0.6b-v2. Previous versions can be [accessed](https://huggingface.co/collections/nvidia/parakeet-659711f49d1469e51546e021) here. 
+<!-- This section describes the evaluation protocols and provides the results. -->
 
-## <span style="color:#466f00;">Training and Evaluation Datasets:</span>
+### Testing Data, Factors & Metrics
 
-### <span style="color:#466f00;">Training</span>
+#### Testing Data
 
-This model was trained using the NeMo toolkit [3], following the strategies below:
+<!-- This should link to a Dataset Card if possible. -->
 
-- Initialized from a wav2vec SSL checkpoint pretrained on the LibriLight dataset[7].  
-- Trained for 150,000 steps on 128 A100 GPUs. 
-- Dataset corpora were balanced using a temperature sampling value of 0.5.  
-- Stage 2 fine-tuning was performed for 2,500 steps on 4 A100 GPUs using approximately 500 hours of high-quality, human-transcribed data of NeMo ASR Set 3.0.  
+[More Information Needed]
 
-Training was conducted using this [example script](https://github.com/NVIDIA/NeMo/blob/main/examples/asr/asr_transducer/speech_to_text_rnnt_bpe.py) and [TDT configuration](https://github.com/NVIDIA/NeMo/blob/main/examples/asr/conf/fastconformer/hybrid_transducer_ctc/fastconformer_hybrid_tdt_ctc_bpe.yaml).
+#### Factors
 
-The tokenizer was constructed from the training set transcripts using this [script](https://github.com/NVIDIA/NeMo/blob/main/scripts/tokenizers/process_asr_text_tokenizer.py).
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
 
-### <span style="color:#466f00;">Training Dataset</span>
-The model was trained on the Granary dataset, consisting of approximately 120,000 hours of English speech data:
+[More Information Needed]
 
-- 10,000 hours from human-transcribed NeMo ASR Set 3.0, including:
-  - LibriSpeech (960 hours)
-  - Fisher Corpus
-  - National Speech Corpus Part 1 
-  - VCTK
-  - VoxPopuli (English)
-  - Europarl-ASR (English)
-  - Multilingual LibriSpeech (MLS English) – 2,000-hour subset
-  - Mozilla Common Voice (v7.0)
-  - AMI
+#### Metrics
 
-- 110,000 hours of pseudo-labeled data from:
-  - YTC (YouTube-Commons) dataset[4]
-  - YODAS dataset [5]
-  - Librilight [7]
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
 
-All transcriptions preserve punctuation and capitalization. The Granary dataset will be made publicly available after presentation at Interspeech 2025.
+[More Information Needed]
 
-**Data Collection Method by dataset**
+### Results
 
-* Hybrid: Automated, Human
+[More Information Needed]
 
-**Labeling Method by dataset**
+#### Summary
 
-* Hybrid: Synthetic, Human 
 
-**Properties:**
 
-* Noise robust data from various sources
-* Single channel, 16kHz sampled data
+## Model Examination [optional]
 
-#### Evaluation Dataset
+<!-- Relevant interpretability work for the model goes here -->
 
-Huggingface Open ASR Leaderboard datasets are used to evaluate the performance of this model. 
+[More Information Needed]
 
-**Data Collection Method by dataset**
-* Human
+## Environmental Impact
 
-**Labeling Method by dataset**
-* Human
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
 
-**Properties:**
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
 
-* All are commonly used for benchmarking English ASR systems.
-* Audio data is typically processed into a 16kHz mono channel format for ASR evaluation, consistent with benchmarks like the [Open ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard).
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
 
-## <span style="color:#466f00;">Performance</span>
+## Technical Specifications [optional]
 
-#### Huggingface Open-ASR-Leaderboard Performance
-The performance of Automatic Speech Recognition (ASR) models is measured using Word Error Rate (WER). Given that this model is trained on a large and diverse dataset spanning multiple domains, it is generally more robust and accurate across various types of audio.
+### Model Architecture and Objective
 
-### Base Performance
-The table below summarizes the WER (%) using a Transducer decoder with greedy decoding (without an external language model):
+[More Information Needed]
 
-| **Model** | **Avg WER** | **AMI** | **Earnings-22** | **GigaSpeech** | **LS test-clean** | **LS test-other** | **SPGI Speech** | **TEDLIUM-v3** | **VoxPopuli** |
-|:-------------|:-------------:|:---------:|:------------------:|:----------------:|:-----------------:|:-----------------:|:------------------:|:----------------:|:---------------:|
-| parakeet-tdt-0.6b-v2 | 6.05 | 11.16 | 11.15 | 9.74 | 1.69 | 3.19 | 2.17 | 3.38 | 5.95 | - |
+### Compute Infrastructure
 
-### Noise Robustness
-Performance across different Signal-to-Noise Ratios (SNR) using MUSAN music and noise samples:
+[More Information Needed]
 
-| **SNR Level** | **Avg WER** | **AMI** | **Earnings** | **GigaSpeech** | **LS test-clean** | **LS test-other** | **SPGI** | **Tedlium** | **VoxPopuli** | **Relative Change** |
-|:---------------|:-------------:|:----------:|:------------:|:----------------:|:-----------------:|:-----------------:|:-----------:|:-------------:|:---------------:|:-----------------:|
-| Clean | 6.05 | 11.16 | 11.15 | 9.74 | 1.69 | 3.19 | 2.17 | 3.38 | 5.95 | - |
-| SNR 50 | 6.04 | 11.11 | 11.12 | 9.74 | 1.70 | 3.18 | 2.18 | 3.34 | 5.98 | +0.25% |
-| SNR 25 | 6.50 | 12.76 | 11.50 | 9.98 | 1.78 | 3.63 | 2.54 | 3.46 | 6.34 | -7.04% |
-| SNR 5 | 8.39 | 19.33 | 13.83 | 11.28 | 2.36 | 5.50 | 3.91 | 3.91 | 6.96 | -38.11% |
+#### Hardware
 
-### Telephony Audio Performance 
-Performance comparison between standard 16kHz audio and telephony-style audio (using μ-law encoding with 16kHz→8kHz→16kHz conversion):
+[More Information Needed]
 
-| **Audio Format** | **Avg WER** | **AMI** | **Earnings** | **GigaSpeech** | **LS test-clean** | **LS test-other** | **SPGI** | **Tedlium** | **VoxPopuli** | **Relative Change** |
-|:-----------------|:-------------:|:----------:|:------------:|:----------------:|:-----------------:|:-----------------:|:-----------:|:-------------:|:---------------:|:-----------------:|
-| Standard 16kHz | 6.05 | 11.16 | 11.15 | 9.74 | 1.69 | 3.19 | 2.17 | 3.38 | 5.95 | - |
-| μ-law 8kHz | 6.32 | 11.98 | 11.16 | 10.02 | 1.78 | 3.52 | 2.20 | 3.38 | 6.52 | -4.10% |
+#### Software
 
-These WER scores were obtained using greedy decoding without an external language model. Additional evaluation details are available on the [Hugging Face ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard).[6]
+[More Information Needed]
 
+## Citation [optional]
 
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
 
-## <span style="color:#466f00;">References</span>
+**BibTeX:**
 
-[1] [Fast Conformer with Linearly Scalable Attention for Efficient Speech Recognition](https://arxiv.org/abs/2305.05084)
+[More Information Needed]
 
-[2] [Efficient Sequence Transduction by Jointly Predicting Tokens and Durations](https://arxiv.org/abs/2304.06795)
+**APA:**
 
-[3] [NVIDIA NeMo Toolkit](https://github.com/NVIDIA/NeMo)
+[More Information Needed]
 
-[4] [Youtube-commons: A massive open corpus for conversational and multimodal data](https://huggingface.co/blog/Pclanglais/youtube-commons)
+## Glossary [optional]
 
-[5] [Yodas: Youtube-oriented dataset for audio and speech](https://arxiv.org/abs/2406.00899)
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
 
-[6] [HuggingFace ASR Leaderboard](https://huggingface.co/spaces/hf-audio/open_asr_leaderboard)
+[More Information Needed]
 
-[7] [MOSEL: 950,000 Hours of Speech Data for Open-Source Speech Foundation Model Training on EU Languages](https://arxiv.org/abs/2410.01036) 
+## More Information [optional]
 
-## <span style="color:#466f00;">Inference:</span>
+[More Information Needed]
 
-**Engine**: 
-* NVIDIA NeMo
+## Model Card Authors [optional]
 
-**Test Hardware**:
-* NVIDIA A10
-* NVIDIA A100
-* NVIDIA A30
-* NVIDIA H100
-* NVIDIA L4
-* NVIDIA L40
-* NVIDIA Turing T4
-* NVIDIA Volta V100
+[More Information Needed]
 
-## <span style="color:#466f00;">Ethical Considerations:</span>
-NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their supporting model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.
+## Model Card Contact
 
-For more detailed information on ethical considerations for this model, please see the Model Card++ Explainability, Bias, Safety & Security, and Privacy Subcards [here](https://developer.nvidia.com/blog/enhancing-ai-transparency-and-ethical-considerations-with-model-card/).
-
-Please report security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).
-
-## <span style="color:#466f00;">Bias:</span>
-
-Field                                                                                               |  Response
----------------------------------------------------------------------------------------------------|---------------
-Participation considerations from adversely impacted groups [protected classes](https://www.senate.ca.gov/content/protected-classes) in model design and testing  |  None
-Measures taken to mitigate against unwanted bias    | None
-
-## <span style="color:#466f00;">Explainability:</span>
-
-Field                                                                                                  |  Response
-------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------
-Intended Domain                                                                   |  Speech to Text Transcription
-Model Type                                                                                            |  FastConformer
-Intended Users                                                                                        |  This model is intended for developers, researchers, academics, and industries building conversational based applications. 
-Output                                                                                                |  Text 
-Describe how the model works                                                                          |  Speech input is encoded into embeddings and passed into conformer-based model and output a text response.
-Name the adversely impacted groups this has been tested to deliver comparable outcomes regardless of  |  Not Applicable
-Technical Limitations & Mitigation                                                                    |  Transcripts may be not 100% accurate. Accuracy varies based on language and characteristics of input audio (Domain, Use Case, Accent, Noise, Speech Type, Context of speech, etc.)
-Verified to have met prescribed NVIDIA quality standards  |  Yes
-Performance Metrics                                                                                   | Word Error Rate
-Potential Known Risks                                                                                 |  If a word is not trained in the language model and not presented in vocabulary, the word is not likely to be recognized. Not recommended for word-for-word/incomplete sentences as accuracy varies based on the context of input text
-Licensing                                                                                             |  GOVERNING TERMS: Use of this model is governed by the [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/legalcode.en) license.
-
-## <span style="color:#466f00;">Privacy:</span>
-
-Field                                                                                                                              |  Response
-----------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------
-Generatable or reverse engineerable personal data?                                                     |  None
-Personal data used to create this model?                                                                                       |  None
-Is there provenance for all datasets used in training?                                                                                |  Yes
-Does data labeling (annotation, metadata) comply with privacy laws?                                                                |  Yes
-Is data compliant with data subject requests for data correction or removal, if such a request was made?                           |  No, not possible with externally-sourced data.
-Applicable Privacy Policy        | https://www.nvidia.com/en-us/about-nvidia/privacy-policy/ 
-
-## <span style="color:#466f00;">Safety:</span>
-
-Field                                               |  Response
----------------------------------------------------|----------------------------------
-Model Application(s)                               |  Speech to Text Transcription
-Describe the life critical impact   |  None
-Use Case Restrictions                              | Abide by [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/legalcode.en) License
-Model and dataset restrictions            |  The Principle of least privilege (PoLP) is applied limiting access for dataset generation and model development. Restrictions enforce dataset access during training, and dataset license constraints adhered to.
+[More Information Needed]