Add model card: Research baseline for Gilbert project with IP clarification

README.md

@@ -1,456 +1,300 @@
 ---
 license: mit
-language: fr
-library_name: transformers
-pipeline_tag: automatic-speech-recognition
-thumbnail: null
 tags:
 - automatic-speech-recognition
-- hf-asr-leaderboard
-datasets:
-- mozilla-foundation/common_voice_13_0
-- facebook/multilingual_librispeech
-- facebook/voxpopuli
-- google/fleurs
-- gigant/african_accented_french
-metrics:
-- wer
-model-index:
-- name: whisper-large-v3-french
-  results:
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: Common Voice 13.0
-      type: mozilla-foundation/common_voice_13_0
-      config: fr
-      split: test
-      args:
-        language: fr
-    metrics:
-    - name: WER
-      type: wer
-      value: 7.28
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: Multilingual LibriSpeech (MLS)
-      type: facebook/multilingual_librispeech
-      config: french
-      split: test
-      args:
-        language: fr
-    metrics:
-    - name: WER
-      type: wer
-      value: 3.98
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: VoxPopuli
-      type: facebook/voxpopuli
-      config: fr
-      split: test
-      args:
-        language: fr
-    metrics:
-    - name: WER
-      type: wer
-      value: 8.91
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: Fleurs
-      type: google/fleurs
-      config: fr_fr
-      split: test
-      args:
-        language: fr
-    metrics:
-    - name: WER
-      type: wer
-      value: 4.84
-  - task:
-      name: Automatic Speech Recognition
-      type: automatic-speech-recognition
-    dataset:
-      name: African Accented French
-      type: gigant/african_accented_french
-      config: fr
-      split: test
-      args:
-        language: fr
-    metrics:
-    - name: WER
-      type: wer
-      value: 4.20
+- asr
+- whisper
+- french
+- speech-recognition
+- stt
+- multilingual
+- research
+- baseline
+library_name: transformers
+pipeline_tag: automatic-speech-recognition
+base_model: openai/whisper-large-v3
 ---
 
-# Whisper-Large-V3-French
-
-Whisper-Large-V3-French is fine-tuned on `openai/whisper-large-v3` to further enhance its performance on the French language. This model has been trained to predict casing, punctuation, and numbers. While this might slightly sacrifice performance, we believe it allows for broader usage.
+# Gilbert-FR-Source — Research Baseline for French Automatic Speech Recognition
 
-This model has been converted into various formats, facilitating its usage across different libraries, including transformers, openai-whisper, fasterwhisper, whisper.cpp, candle, mlx, etc.
+## Overview
 
-## Table of Contents
+**Gilbert-FR-Source** is the foundational baseline model for the **Gilbert research project**, a comprehensive initiative focused on developing state-of-the-art automatic speech recognition (ASR) systems optimized for French language applications. This model serves as the **frozen reference point** for all subsequent research, fine-tuning, and development work within the Gilbert ecosystem.
 
-- [Performance](#performance)
-- [Usage](#usage)
-    - [Hugging Face Pipeline](#hugging-face-pipeline)
-    - [Hugging Face Low-level APIs](#hugging-face-low-level-apis)
-    - [Speculative Decoding](#speculative-decoding)
-    - [OpenAI Whisper](#openai-whisper)
-    - [Faster Whisper](#faster-whisper)
-    - [Whisper.cpp](#whispercpp)
-    - [Candle](#candle)
-    - [MLX](#mlx)
-- [Training details](#training-details)
-- [Acknowledgements](#acknowledgements)
+**Important Notice on Intellectual Property:**
+- This baseline model (`MEscriva/gilbert-fr-source`) is distributed under the MIT License, allowing research and commercial use.
+- **All derivative models, fine-tuned variants, and specialized models developed from this baseline as part of the Gilbert project are the exclusive intellectual property of Lexia France.**
+- While this baseline can be used freely under MIT terms, any models built upon it for the Gilbert project are proprietary and subject to separate licensing terms.
 
-## Performance
+---
 
-We evaluated our model on both short and long-form transcriptions, and also tested it on both in-distribution and out-of-distribution datasets to conduct a comprehensive analysis assessing its accuracy, generalizability, and robustness.
+## Research Context
 
-Please note that the reported WER is the result after converting numbers to text, removing punctuation (except for apostrophes and hyphens), and converting all characters to lowercase.
+The Gilbert project is a systematic research and development effort aimed at creating highly specialized ASR systems for:
 
-All evaluation results on the public datasets can be found [here](https://drive.google.com/drive/folders/1rFIh6yXRVa9RZ0ieZoKiThFZgQ4STPPI?usp=drive_link).
+- **Professional meeting transcription** (hybrid and remote meetings)
+- **Long-form multi-speaker discourse** (30-120 minute sessions)
+- **Institutional environments** (education, public sector, healthcare)
+- **Constrained audio conditions** (telephony, VoIP, low signal-to-noise ratio)
+- **Sociolinguistic diversity** (African, Canadian, Belgian, and other French accents)
 
-### Short-Form Transcription
+This baseline model provides the **controlled starting point** for all experimental work, ensuring reproducibility and enabling fair comparison across different research directions.
 
-![eval-short-form](https://huggingface.co/bofenghuang/whisper-large-v3-french/resolve/main/assets/whisper_fr_eval_short_form.png)
+---
 
-Due to the lack of readily available out-of-domain (OOD) and long-form test sets in French, we evaluated using internal test sets from [Zaion Lab](https://zaion.ai/). These sets comprise human-annotated audio-transcription pairs from call center conversations, which are notable for their significant background noise and domain-specific terminology.
+## Model Details
 
-### Long-Form Transcription
+### Architecture
 
-![eval-long-form](https://huggingface.co/bofenghuang/whisper-large-v3-french/resolve/main/assets/whisper_fr_eval_long_form.png)
+- **Base Model:** OpenAI Whisper Large V3
+- **Fine-tuning:** Optimized for French language performance
+- **Framework:** Compatible with Hugging Face Transformers, OpenAI Whisper, CTranslate2, ONNX Runtime, and MLX
+- **Model Size:** ~3.2 GB (full precision)
 
-The long-form transcription was run using the 🤗 Hugging Face pipeline for quicker evaluation. Audio files were segmented into 30-second chunks and processed in parallel.
+### Key Characteristics
 
-## Usage
+- **Language:** French (primary), with multilingual capabilities
+- **Context Length:** Long-form audio support (up to 30 minutes per segment)
+- **Output:** Text transcription with word-level timestamps
+- **Performance:** Optimized for French speech recognition accuracy
 
-### Hugging Face Pipeline
+---
 
-The model can easily used with the 🤗 Hugging Face [`pipeline`](https://huggingface.co/docs/transformers/main_classes/pipelines#transformers.AutomaticSpeechRecognitionPipeline) class for audio transcription.
+## Intended Use
 
-For long-form transcription (> 30 seconds), you can activate the process by passing the `chunk_length_s` argument. This approach segments the audio into smaller segments, processes them in parallel, and then joins them at the strides by finding the longest common sequence. While this chunked long-form approach may have a slight compromise in performance compared to OpenAI's sequential algorithm, it provides 9x faster inference speed.
+### Research and Development
 
-```python
-import torch
-from datasets import load_dataset
-from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor, pipeline
+This model is intended for:
 
-device = "cuda:0" if torch.cuda.is_available() else "cpu"
-torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+1. **Research Baseline:** Use as a reference point for ASR research and experimentation
+2. **Comparative Studies:** Benchmark against this baseline when evaluating new architectures or training strategies
+3. **Fine-tuning Foundation:** Use as a starting point for domain-specific fine-tuning (subject to Gilbert project IP terms)
+4. **Educational Purposes:** Learning and understanding ASR model behavior
 
-# Load model
-model_name_or_path = "bofenghuang/whisper-large-v3-french"
-processor = AutoProcessor.from_pretrained(model_name_or_path)
-model = AutoModelForSpeechSeq2Seq.from_pretrained(
-    model_name_or_path,
-    torch_dtype=torch_dtype,
-    low_cpu_mem_usage=True,
-)
-model.to(device)
+### Production Use
 
-# Init pipeline
-pipe = pipeline(
-    "automatic-speech-recognition",
-    model=model,
-    feature_extractor=processor.feature_extractor,
-    tokenizer=processor.tokenizer,
-    torch_dtype=torch_dtype,
-    device=device,
-    # chunk_length_s=30,  # for long-form transcription
-    max_new_tokens=128,
-)
+While this baseline model can be used directly, **production deployments should use specialized Gilbert models** that are optimized for specific use cases and domains. Contact the Gilbert team for production-grade models.
 
-# Example audio
-dataset = load_dataset("bofenghuang/asr-dummy", "fr", split="test")
-sample = dataset[0]["audio"]
+---
 
-# Run pipeline
-result = pipe(sample)
-print(result["text"])
-```
+## Performance Benchmarks
 
-### Hugging Face Low-level APIs
+### Reference Results
 
-You can also use the 🤗 Hugging Face low-level APIs for transcription, offering greater control over the process, as demonstrated below:
+The following WER (Word Error Rate) scores serve as **baseline reference** for future Gilbert model development:
 
-```python
-import torch
-from datasets import load_dataset
-from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor
+| Dataset | WER | Notes |
+|---------|-----|-------|
+| MLS (FR) | 3.98% | Multilingual LibriSpeech French |
+| Common Voice FR (v13.0) | 7.28% | Diverse French speech |
+| VoxPopuli (FR) | 8.91% | European Parliament speeches |
+| Fleurs (FR) | 4.84% | FLORES evaluation |
+| African Accented French | 4.20% | Regional accent evaluation |
 
-device = "cuda:0" if torch.cuda.is_available() else "cpu"
-torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+**Note:** These results represent the **upper bound** before targeted fine-tuning. Future Gilbert variants will be evaluated against these baselines to measure improvement.
 
-# Load model
-model_name_or_path = "bofenghuang/whisper-large-v3-french"
-processor = AutoProcessor.from_pretrained(model_name_or_path)
-model = AutoModelForSpeechSeq2Seq.from_pretrained(
-    model_name_or_path,
-    torch_dtype=torch_dtype,
-    low_cpu_mem_usage=True,
-)
-model.to(device)
-
-# Example audio
-dataset = load_dataset("bofenghuang/asr-dummy", "fr", split="test")
-sample = dataset[0]["audio"]
-
-# Extract feautres
-input_features = processor(
-    sample["array"], sampling_rate=sample["sampling_rate"], return_tensors="pt"
-).input_features
+---
 
+## Usage
 
-# Generate tokens
-predicted_ids = model.generate(
-    input_features.to(dtype=torch_dtype).to(device), max_new_tokens=128
-)
+### Installation
 
-# Detokenize to text
-transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0]
-print(transcription)
+```bash
+pip install transformers torch torchaudio librosa soundfile
 ```
 
-### Speculative Decoding
-
-[Speculative decoding](https://huggingface.co/blog/whisper-speculative-decoding) can be achieved using a draft model, essentially a distilled version of Whisper. This approach guarantees identical outputs to using the main Whisper model alone, offers a 2x faster inference speed, and incurs only a slight increase in memory overhead. 
-
-Since the distilled Whisper has the same encoder as the original, only its decoder need to be loaded, and encoder outputs are shared between the main and draft models during inference.
-
-Using speculative decoding with the Hugging Face pipeline is simple - just specify the `assistant_model` within the generation configurations.
+### Basic Usage with Transformers
 
 ```python
+from transformers import AutoModelForSpeechSeq2Seq, AutoProcessor
 import torch
-from datasets import load_dataset
-from transformers import (
-    AutoModelForCausalLM,
-    AutoModelForSpeechSeq2Seq,
-    AutoProcessor,
-    pipeline,
-)
 
-device = "cuda:0" if torch.cuda.is_available() else "cpu"
-torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+model_id = "MEscriva/gilbert-fr-source"
+device = "cuda" if torch.cuda.is_available() else "cpu"
+torch_dtype = torch.float16 if device == "cuda" else torch.float32
 
-# Load model
-model_name_or_path = "bofenghuang/whisper-large-v3-french"
-processor = AutoProcessor.from_pretrained(model_name_or_path)
+processor = AutoProcessor.from_pretrained(model_id)
 model = AutoModelForSpeechSeq2Seq.from_pretrained(
-    model_name_or_path,
+    model_id,
     torch_dtype=torch_dtype,
-    low_cpu_mem_usage=True,
+    low_cpu_mem_usage=True
 )
 model.to(device)
 
-# Load draft model
-assistant_model_name_or_path = "bofenghuang/whisper-large-v3-french-distil-dec2"
-assistant_model = AutoModelForCausalLM.from_pretrained(
-    assistant_model_name_or_path,
-    torch_dtype=torch_dtype,
-    low_cpu_mem_usage=True,
-)
-assistant_model.to(device)
-
-# Init pipeline
-pipe = pipeline(
-    "automatic-speech-recognition",
-    model=model,
-    feature_extractor=processor.feature_extractor,
-    tokenizer=processor.tokenizer,
-    torch_dtype=torch_dtype,
-    device=device,
-    generate_kwargs={"assistant_model": assistant_model},
-    max_new_tokens=128,
-)
-
-# Example audio
-dataset = load_dataset("bofenghuang/asr-dummy", "fr", split="test")
-sample = dataset[0]["audio"]
-
-# Run pipeline
-result = pipe(sample)
-print(result["text"])
-```
-
-### OpenAI Whisper
-
-You can also employ the sequential long-form decoding algorithm with a sliding window and temperature fallback, as outlined by OpenAI in their original [paper](https://arxiv.org/abs/2212.04356).
-
-First, install the [openai-whisper](https://github.com/openai/whisper) package:
-
-```bash
-pip install -U openai-whisper
-```
-
-Then, download the converted model:
-
-```bash
-python -c "from huggingface_hub import hf_hub_download; hf_hub_download(repo_id='bofenghuang/whisper-large-v3-french', filename='original_model.pt', local_dir='./models/whisper-large-v3-french')"
+# Process audio
+audio_path = "your_audio.wav"
+inputs = processor(audio_path, return_tensors="pt", sampling_rate=16000)
+inputs = {k: v.to(device) for k, v in inputs.items()}
+
+with torch.no_grad():
+    generated_ids = model.generate(
+        inputs["input_features"],
+        language="fr",
+        task="transcribe"
+    )
+
+transcription = processor.batch_decode(
+    generated_ids,
+    skip_special_tokens=True
+)[0]
 ```
 
-Now, you can transcirbe audio files by following the usage instructions provided in the repository:
+### Usage with OpenAI Whisper
 
 ```python
 import whisper
-from datasets import load_dataset
 
-# Load model
-model = whisper.load_model("./models/whisper-large-v3-french/original_model.pt")
+# Load the model
+model = whisper.load_model("large-v3")
 
-# Example audio
-dataset = load_dataset("bofenghuang/asr-dummy", "fr", split="test")
-sample = dataset[0]["audio"]["array"].astype("float32")
+# Transcribe French audio
+result = model.transcribe(
+    "audio.wav",
+    language="fr",
+    task="transcribe"
+)
 
-# Transcribe
-result = model.transcribe(sample, language="fr")
 print(result["text"])
 ```
 
-### Faster Whisper
+---
 
-Faster Whisper is a reimplementation of OpenAI's Whisper models and the sequential long-form decoding algorithm in the [CTranslate2](https://github.com/OpenNMT/CTranslate2) format.
+## Research Methodology
 
-Compared to openai-whisper, it offers up to 4x faster inference speed, while consuming less memory. Additionally, the model can be quantized into int8, further enhancing its efficiency on both CPU and GPU.
+### Baseline Purpose
 
-First, install the [faster-whisper](https://github.com/SYSTRAN/faster-whisper) package:
+This model serves as:
 
-```bash
-pip install faster-whisper
-```
+1. **Frozen Reference:** Weights remain unchanged to ensure consistent baseline comparisons
+2. **Reproducibility Anchor:** All experiments reference this exact checkpoint
+3. **Version Control:** Future Gilbert models explicitly reference this baseline version for traceability
 
-Then, download the model converted to the CTranslate2 format:
+### Evaluation Standards
 
-```bash
-python -c "from huggingface_hub import snapshot_download; snapshot_download(repo_id='bofenghuang/whisper-large-v3-french', local_dir='./models/whisper-large-v3-french', allow_patterns='ctranslate2/*')"
-```
+- **WER Calculation:** Standard normalization (lowercasing, punctuation removal)
+- **Metrics:** Word Error Rate (WER), Character Error Rate (CER), BLEU score
+- **Advanced Metrics:** Speaker-attributed WER (SA-WER), long-context stability (internal research)
 
-Now, you can transcirbe audio files by following the usage instructions provided in the repository:
+### Versioning
 
-```python
-from datasets import load_dataset
-from faster_whisper import WhisperModel
+- **Current Version:** 0.1 (Research Baseline)
+- **Future Versions:** All Gilbert model variants will reference this baseline version
 
-# Load model
-model = WhisperModel("./models/whisper-large-v3-french/ctranslate2", device="cuda", compute_type="float16")  # Run on GPU with FP16
+---
 
-# Example audio
-dataset = load_dataset("bofenghuang/asr-dummy", "fr", split="test")
-sample = dataset[0]["audio"]["array"].astype("float32")
+## Limitations
 
-segments, info = model.transcribe(sample, beam_size=5, language="fr")
+This baseline model inherits known limitations from Whisper and the underlying training data:
 
-for segment in segments:
-    print("[%.2fs -> %.2fs] %s" % (segment.start, segment.end, segment.text))
-```
+1. **Overlapping Speech:** Sensitivity to simultaneous speakers
+2. **Long-form Decoding:** Occasional hallucinations in very long audio segments
+3. **Domain Shift:** Suboptimal performance on spontaneous dialogue without fine-tuning
+4. **Accent Distribution:** Potential biases related to accent representation in training data
+5. **Telephony Bandwidth:** Suboptimal performance on narrowband (8 kHz) audio without adaptation
 
-### Whisper.cpp
+**Understanding and quantifying these limitations is a core objective of the Gilbert research roadmap.**
 
-Whisper.cpp is a reimplementation of OpenAI's Whisper models, crafted in plain C/C++ without any dependencies. It offers compatibility with various backends and platforms.
+---
 
-Additionally, the model can be quantized to either 4-bit or 5-bit integers, further enhancing its efficiency.
+## Future Research Directions
 
-First, clone and build the [whisper.cpp](https://github.com/ggerganov/whisper.cpp) repository:
+The following specialized models will be developed as independent checkpoints from this baseline:
 
-```bash
-git clone https://github.com/ggerganov/whisper.cpp.git
-cd whisper.cpp
+### Planned Gilbert Models
 
-# build the main example
-make
-```
+1. **Gilbert-FR-Longform-v1**
+   - Optimized for long meetings (30-120 minutes)
+   - Multi-speaker interaction handling
+   - Discourse-level context stability
 
-Next, download the converted ggml weights from the Hugging Face Hub:
+2. **Gilbert-FR-Accents-v1**
+   - Robustness to regional and international French accents
+   - African, Canadian, Belgian accent optimization
 
-```bash
-# Download model quantized with Q5_0 method
-python -c "from huggingface_hub import hf_hub_download; hf_hub_download(repo_id='bofenghuang/whisper-large-v3-french', filename='ggml-model-q5_0.bin', local_dir='./models/whisper-large-v3-french')"
-```
+3. **Gilbert-FR-Telephone-v1**
+   - Optimized for 8 kHz VoIP/call-center speech
+   - Narrowband audio adaptation
 
-Now, you can transcribe an audio file using the following command:
+4. **Gilbert-Multilingual-v1**
+   - Extended cross-lingual performance
+   - Optimized French anchors with multilingual support
 
-```bash
-./main -m ./models/whisper-large-v3-french/ggml-model-q5_0.bin -l fr -f /path/to/audio/file --print-colors
-```
+**All future Gilbert models are the exclusive intellectual property of Lexia France** and will include detailed evaluation reports adhering to research reproducibility standards.
 
-### Candle
+---
 
-[Candle-whisper](https://github.com/huggingface/candle/tree/main/candle-examples/examples/whisper) is a reimplementation of OpenAI's Whisper models in the candle format - a lightweight ML framework built in Rust.
+## Intellectual Property and Licensing
 
-First, clone the [candle](https://github.com/huggingface/candle) repository:
+### License for This Baseline
 
-```bash
-git clone https://github.com/huggingface/candle.git
-cd candle/candle-examples/examples/whisper
-```
+This baseline model (`MEscriva/gilbert-fr-source`) is distributed under the **MIT License**, allowing:
 
-Transcribe an audio file using the following command:
+- ✅ Commercial use
+- ✅ Modification
+- ✅ Distribution
+- ✅ Private use
+- ✅ Patent use
 
-```bash
-cargo run --example whisper --release -- --model large-v3 --model-id bofenghuang/whisper-large-v3-french --language fr --input /path/to/audio/file
-```
+See the `LICENSE` file for full terms.
 
-In order to use CUDA add `--features cuda` to the example command line:
+### Intellectual Property Notice
 
-```bash
-cargo run --example whisper --release --features cuda -- --model large-v3 --model-id bofenghuang/whisper-large-v3-french --language fr --input /path/to/audio/file
-```
+**Important:** While this baseline model is available under MIT License:
 
-### MLX
+- **All derivative models, fine-tuned variants, and specialized models developed as part of the Gilbert project are the exclusive intellectual property of Lexia France.**
+- Use of this baseline for Gilbert project development implies acceptance of these IP terms.
+- Commercial use of Gilbert project derivatives requires separate licensing agreements.
 
-[MLX-Whisper](https://github.com/ml-explore/mlx-examples/tree/main/whisper) is a reimplementation of OpenAI's Whisper models in the [MLX](https://github.com/ml-explore/mlx) format - a ML framework on Apple silicon. It supports features like lazy computation, unified memory management, etc.
+For licensing inquiries regarding Gilbert project models, contact: **mathis@lexiapro.fr**
 
-First, clone the [MLX Examples](https://github.com/ml-explore/mlx-examples) repository:
+---
 
-```bash
-git clone https://github.com/ml-explore/mlx-examples.git
-cd mlx-examples/whisper
-```
+## Citation
 
-Next, install the dependencies:
+If you use this baseline model in your research, please cite:
 
-```bash
-pip install -r requirements.txt
+```bibtex
+@software{gilbert_fr_source_2024,
+  title={Gilbert-FR-Source: Research Baseline for French Automatic Speech Recognition},
+  author={MEscriva and Lexia France},
+  year={2024},
+  url={https://huggingface.co/MEscriva/gilbert-fr-source},
+  version={0.1},
+  note={Research baseline for the Gilbert project}
+}
 ```
 
-Download the pytorch checkpoint in the original OpenAI format and convert it into MLX format (We haven't included the converted version here since the repository is already heavy and the conversion is very fast):
+---
 
-```bash
-# Download
-python -c "from huggingface_hub import hf_hub_download; hf_hub_download(repo_id='bofenghuang/whisper-large-v3-french', filename='original_model.pt', local_dir='./models/whisper-large-v3-french')"
-# Convert into .npz
-python convert.py --torch-name-or-path ./models/whisper-large-v3-french/original_model.pt --mlx-path ./mlx_models/whisper-large-v3-french
-```
+## Acknowledgments
 
-Now, you can transcribe audio with:
+This baseline model is based on:
+- **OpenAI Whisper Large V3** (MIT License)
+- **bofenghuang/whisper-large-v3-french** (French fine-tuning)
 
-```python
-import whisper
+We acknowledge the contributions of the open-source community and the original Whisper research team.
 
-result = whisper.transcribe("/path/to/audio/file", path_or_hf_repo="mlx_models/whisper-large-v3-french", language="fr")
-print(result["text"])
-```
+---
+
+## Contact
 
-## Training details
+For research collaboration, evaluation access, or technical inquiries:
 
-We've collected a composite dataset consisting of over 2,500 hours of French speech recognition data, which incldues datasets such as [Common Voice 13.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_13_0), [Multilingual LibriSpeech](https://huggingface.co/datasets/facebook/multilingual_librispeech), [Voxpopuli](https://huggingface.co/datasets/facebook/voxpopuli), [Fleurs](https://huggingface.co/datasets/google/fleurs), [Multilingual TEDx](https://www.openslr.org/100/), [MediaSpeech](https://www.openslr.org/108/), [African Accented French](https://huggingface.co/datasets/gigant/african_accented_french), etc.
+- **Website:** [https://gilbert-assistant.fr](https://gilbert-assistant.fr)
+- **Email:** mathis@lexiapro.fr
+- **Repository:** [https://huggingface.co/MEscriva/gilbert-fr-source](https://huggingface.co/MEscriva/gilbert-fr-source)
 
-Given that some datasets, like MLS, only offer text without case or punctuation, we employed a customized version of 🤗 [Speechbox](https://github.com/huggingface/speechbox) to restore case and punctuation from a limited set of symbols using the [bofenghuang/whisper-large-v2-cv11-french](bofenghuang/whisper-large-v2-cv11-french) model.
+---
 
-However, even within these datasets, we observed certain quality issues. These ranged from mismatches between audio and transcription in terms of language or content, poorly segmented utterances, to missing words in scripted speech, etc. We've built a pipeline to filter out many of these problematic utterances, aiming to enhance the dataset's quality. As a result, we excluded more than 10% of the data, and when we retrained the model, we noticed a significant reduction of hallucination.
+## Changelog
 
-For training, we employed the [script](https://github.com/huggingface/transformers/blob/main/examples/pytorch/speech-recognition/run_speech_recognition_seq2seq.py) available in the 🤗 Transformers repository. The model training took place on the [Jean-Zay supercomputer](http://www.idris.fr/eng/jean-zay/jean-zay-presentation-eng.html) at GENCI, and we extend our gratitude to the IDRIS team for their responsive support throughout the project.
+### Version 0.1 (2024-12-19)
+- Initial research baseline release
+- Based on Whisper Large V3 with French optimization
+- Established as frozen reference point for Gilbert project
+- Documentation of baseline performance metrics
+
+---
 
-## Acknowledgements
+**© 2024 Lexia France. All rights reserved for Gilbert project derivatives.**
 
-- OpenAI for creating and open-sourcing the [Whisper model](https://arxiv.org/abs/2212.04356)
-- 🤗 Hugging Face for integrating the Whisper model and providing the training codebase within the [Transformers](https://github.com/huggingface/transformers) repository
-- [Genci](https://genci.fr/) for their generous contribution of GPU hours to this project