Upload folder using huggingface_hub

.gitattributes

@@ -1,35 +1,27 @@
 *.7z filter=lfs diff=lfs merge=lfs -text
 *.arrow filter=lfs diff=lfs merge=lfs -text
 *.bin filter=lfs diff=lfs merge=lfs -text
+*.bin.* filter=lfs diff=lfs merge=lfs -text
 *.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
 *.ftz filter=lfs diff=lfs merge=lfs -text
 *.gz filter=lfs diff=lfs merge=lfs -text
 *.h5 filter=lfs diff=lfs merge=lfs -text
 *.joblib filter=lfs diff=lfs merge=lfs -text
 *.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
 *.model filter=lfs diff=lfs merge=lfs -text
 *.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
 *.onnx filter=lfs diff=lfs merge=lfs -text
 *.ot filter=lfs diff=lfs merge=lfs -text
 *.parquet filter=lfs diff=lfs merge=lfs -text
 *.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
 *.pt filter=lfs diff=lfs merge=lfs -text
 *.pth filter=lfs diff=lfs merge=lfs -text
 *.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text 
 *.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
 *.tflite filter=lfs diff=lfs merge=lfs -text
 *.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
 *.xz filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
+*.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,148 @@
+---
+language:
+- fr
+license: apache-2.0
+tags:
+- automatic-speech-recognition
+- fr
+- hf-asr-leaderboard
+- mozilla-foundation/common_voice_8_0
+- robust-speech-event
+datasets:
+- mozilla-foundation/common_voice_8_0
+model-index:
+- name: XLS-R Wav2Vec2 French by Jonatas Grosman
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: Common Voice 8
+      type: mozilla-foundation/common_voice_8_0
+      args: fr
+    metrics:
+    - name: Test WER
+      type: wer
+      value: 16.85
+    - name: Test CER
+      type: cer
+      value: 4.66
+    - name: Test WER (+LM)
+      type: wer
+      value: 16.32
+    - name: Test CER (+LM)
+      type: cer
+      value: 4.21
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: Robust Speech Event - Dev Data
+      type: speech-recognition-community-v2/dev_data
+      args: fr
+    metrics:
+    - name: Dev WER
+      type: wer
+      value: 22.34
+    - name: Dev CER
+      type: cer
+      value: 9.88
+    - name: Dev WER (+LM)
+      type: wer
+      value: 17.16
+    - name: Dev CER (+LM)
+      type: cer
+      value: 9.38
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: Robust Speech Event - Test Data
+      type: speech-recognition-community-v2/eval_data
+      args: fr
+    metrics:
+    - name: Test WER
+      type: wer
+      value: 19.15
+---
+
+# Fine-tuned XLS-R 1B model for speech recognition in French
+
+Fine-tuned [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on French using the train and validation splits of [Common Voice 8.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_8_0), [MediaSpeech](https://www.openslr.org/108/), [Multilingual TEDx](http://www.openslr.org/100), [Multilingual LibriSpeech](https://www.openslr.org/94/), and [Voxpopuli](https://github.com/facebookresearch/voxpopuli).
+When using this model, make sure that your speech input is sampled at 16kHz.
+
+This model has been fine-tuned by the [HuggingSound](https://github.com/jonatasgrosman/huggingsound) tool, and thanks to the GPU credits generously given by the [OVHcloud](https://www.ovhcloud.com/en/public-cloud/ai-training/) :)
+
+## Usage
+
+Using the [HuggingSound](https://github.com/jonatasgrosman/huggingsound) library:
+
+```python
+from huggingsound import SpeechRecognitionModel
+
+model = SpeechRecognitionModel("jonatasgrosman/wav2vec2-xls-r-1b-french")
+audio_paths = ["/path/to/file.mp3", "/path/to/another_file.wav"]
+
+transcriptions = model.transcribe(audio_paths)
+```
+
+Writing your own inference script:
+
+```python
+import torch
+import librosa
+from datasets import load_dataset
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
+
+LANG_ID = "fr"
+MODEL_ID = "jonatasgrosman/wav2vec2-xls-r-1b-french"
+SAMPLES = 10
+
+test_dataset = load_dataset("common_voice", LANG_ID, split=f"test[:{SAMPLES}]")
+
+processor = Wav2Vec2Processor.from_pretrained(MODEL_ID)
+model = Wav2Vec2ForCTC.from_pretrained(MODEL_ID)
+
+# Preprocessing the datasets.
+# We need to read the audio files as arrays
+def speech_file_to_array_fn(batch):
+    speech_array, sampling_rate = librosa.load(batch["path"], sr=16_000)
+    batch["speech"] = speech_array
+    batch["sentence"] = batch["sentence"].upper()
+    return batch
+
+test_dataset = test_dataset.map(speech_file_to_array_fn)
+inputs = processor(test_dataset["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
+
+with torch.no_grad():
+    logits = model(inputs.input_values, attention_mask=inputs.attention_mask).logits
+
+predicted_ids = torch.argmax(logits, dim=-1)
+predicted_sentences = processor.batch_decode(predicted_ids)
+```
+
+## Evaluation Commands
+
+1. To evaluate on `mozilla-foundation/common_voice_8_0` with split `test`
+
+```bash
+python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-french --dataset mozilla-foundation/common_voice_8_0 --config fr --split test
+```
+
+2. To evaluate on `speech-recognition-community-v2/dev_data`
+
+```bash
+python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-french --dataset speech-recognition-community-v2/dev_data --config fr --split validation --chunk_length_s 5.0 --stride_length_s 1.0
+```
+
+## Citation
+If you want to cite this model you can use this:
+
+```bibtex
+@misc{grosman2021xlsr-1b-french,
+  title={Fine-tuned {XLS-R} 1{B} model for speech recognition in {F}rench},
+  author={Grosman, Jonatas},
+  howpublished={\url{https://huggingface.co/jonatasgrosman/wav2vec2-xls-r-1b-french}},
+  year={2022}
+}
+```

alphabet.json

@@ -0,0 +1 @@
+{"labels": ["", "<s>", "</s>", "\u2047", " ", "'", "-", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", "\u00e0", "\u00e2", "\u00e3", "\u00e7", "\u00e8", "\u00e9", "\u00ea", "\u00eb", "\u00ee", "\u00ef", "\u00f4", "\u00f9", "\u00fb", "\u0153"], "is_bpe": false}

config.json

@@ -0,0 +1,107 @@
+{
+  "_name_or_path": "facebook/wav2vec2-xls-r-1b",
+  "activation_dropout": 0.05,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.05,
+  "bos_token_id": 1,
+  "classifier_proj_size": 256,
+  "codevector_dim": 1024,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": false,
+  "diversity_loss_weight": 0.1,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.05,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.05,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.05,
+  "hidden_size": 1280,
+  "initializer_range": 0.02,
+  "intermediate_size": 5120,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.05,
+  "mask_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
+  "model_type": "wav2vec2",
+  "num_adapter_layers": 3,
+  "num_attention_heads": 16,
+  "num_codevector_groups": 2,
+  "num_codevectors_per_group": 320,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 48,
+  "num_negatives": 100,
+  "output_hidden_size": 1280,
+  "pad_token_id": 0,
+  "proj_codevector_dim": 1024,
+  "tdnn_dilation": [
+    1,
+    2,
+    3,
+    1,
+    1
+  ],
+  "tdnn_dim": [
+    512,
+    512,
+    512,
+    512,
+    1500
+  ],
+  "tdnn_kernel": [
+    5,
+    3,
+    3,
+    1,
+    1
+  ],
+  "torch_dtype": "float32",
+  "transformers_version": "4.16.0.dev0",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 47,
+  "xvector_output_dim": 512
+}

eval.py

@@ -0,0 +1,164 @@
+#!/usr/bin/env python3
+from datasets import load_dataset, load_metric, Audio, Dataset
+from transformers import pipeline, AutoFeatureExtractor, AutoTokenizer, AutoConfig, AutoModelForCTC, Wav2Vec2Processor, Wav2Vec2ProcessorWithLM
+import re
+import torch
+import argparse
+from typing import Dict
+
+def log_results(result: Dataset, args: Dict[str, str]):
+    """ DO NOT CHANGE. This function computes and logs the result metrics. """
+
+    log_outputs = args.log_outputs
+    dataset_id = "_".join(args.dataset.split("/") + [args.config, args.split])
+
+    # load metric
+    wer = load_metric("wer")
+    cer = load_metric("cer")
+
+    # compute metrics
+    wer_result = wer.compute(references=result["target"], predictions=result["prediction"])
+    cer_result = cer.compute(references=result["target"], predictions=result["prediction"])
+
+    # print & log results
+    result_str = (
+        f"WER: {wer_result}\n"
+        f"CER: {cer_result}"
+    )
+    print(result_str)
+
+    with open(f"{dataset_id}_eval_results.txt", "w") as f:
+        f.write(result_str)
+
+    # log all results in text file. Possibly interesting for analysis
+    if log_outputs is not None:
+        pred_file = f"log_{dataset_id}_predictions.txt"
+        target_file = f"log_{dataset_id}_targets.txt"
+
+        with open(pred_file, "w") as p, open(target_file, "w") as t:
+
+            # mapping function to write output
+            def write_to_file(batch, i):
+                p.write(f"{i}" + "\n")
+                p.write(batch["prediction"] + "\n")
+                t.write(f"{i}" + "\n")
+                t.write(batch["target"] + "\n")
+
+            result.map(write_to_file, with_indices=True)
+
+
+def normalize_text(text: str, invalid_chars_regex: str, to_lower: bool) -> str:
+    """ DO ADAPT FOR YOUR USE CASE. this function normalizes the target text. """
+
+    text = text.lower() if to_lower else text.upper()
+
+    text = re.sub(invalid_chars_regex, " ", text)
+
+    text = re.sub("\s+", " ", text).strip()
+
+    return text
+
+
+def main(args):
+    # load dataset
+    dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
+
+    # for testing: only process the first two examples as a test
+    # dataset = dataset.select(range(10))
+
+    # load processor
+    if args.greedy:
+        processor = Wav2Vec2Processor.from_pretrained(args.model_id)
+        decoder = None
+    else:
+        processor = Wav2Vec2ProcessorWithLM.from_pretrained(args.model_id)
+        decoder = processor.decoder
+
+    feature_extractor = processor.feature_extractor
+    tokenizer = processor.tokenizer
+
+    # resample audio
+    dataset = dataset.cast_column("audio", Audio(sampling_rate=feature_extractor.sampling_rate))
+
+    # load eval pipeline
+    if args.device is None:
+        args.device = 0 if torch.cuda.is_available() else -1
+    
+    config = AutoConfig.from_pretrained(args.model_id)
+    model = AutoModelForCTC.from_pretrained(args.model_id)
+    
+    #asr = pipeline("automatic-speech-recognition", model=args.model_id, device=args.device)
+    asr = pipeline("automatic-speech-recognition", config=config, model=model, tokenizer=tokenizer, 
+                   feature_extractor=feature_extractor, decoder=decoder, device=args.device)
+    
+    # build normalizer config
+    tokenizer = AutoTokenizer.from_pretrained(args.model_id)
+    tokens = [x for x in tokenizer.convert_ids_to_tokens(range(0, tokenizer.vocab_size))]
+    special_tokens = [
+        tokenizer.pad_token, tokenizer.word_delimiter_token,
+        tokenizer.unk_token, tokenizer.bos_token,
+        tokenizer.eos_token,
+    ]
+    non_special_tokens = [x for x in tokens if x not in special_tokens]
+    invalid_chars_regex = f"[^\s{re.escape(''.join(set(non_special_tokens)))}]"
+    normalize_to_lower = False
+    for token in non_special_tokens:
+        if token.isalpha() and token.islower():
+            normalize_to_lower = True
+            break
+
+    # map function to decode audio
+    def map_to_pred(batch, args=args, asr=asr, invalid_chars_regex=invalid_chars_regex, normalize_to_lower=normalize_to_lower):
+        prediction = asr(batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s)
+
+        batch["prediction"] = prediction["text"]
+        batch["target"] = normalize_text(batch["sentence"], invalid_chars_regex, normalize_to_lower)
+        return batch
+
+    # run inference on all examples
+    result = dataset.map(map_to_pred, remove_columns=dataset.column_names)
+
+    # filtering out empty targets
+    result = result.filter(lambda example: example["target"] != "")
+
+    # compute and log_results
+    # do not change function below
+    log_results(result, args)
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument(
+        "--model_id", type=str, required=True, help="Model identifier. Should be loadable with 🤗 Transformers"
+    )
+    parser.add_argument(
+        "--dataset", type=str, required=True, help="Dataset name to evaluate the `model_id`. Should be loadable with 🤗 Datasets"
+    )
+    parser.add_argument(
+        "--config", type=str, required=True, help="Config of the dataset. *E.g.* `'en'`  for Common Voice"
+    )
+    parser.add_argument(
+        "--split", type=str, required=True, help="Split of the dataset. *E.g.* `'test'`"
+    )
+    parser.add_argument(
+        "--chunk_length_s", type=float, default=None, help="Chunk length in seconds. Defaults to None. For long audio files a good value would be 5.0 seconds."
+    )
+    parser.add_argument(
+        "--stride_length_s", type=float, default=None, help="Stride of the audio chunks. Defaults to None. For long audio files a good value would be 1.0 seconds."
+    )
+    parser.add_argument(
+        "--log_outputs", action='store_true', help="If defined, write outputs to log file for analysis."
+    )
+    parser.add_argument(
+        "--greedy", action='store_true', help="If defined, the LM will be ignored during inference."
+    )
+    parser.add_argument(
+        "--device",
+        type=int,
+        default=None,
+        help="The device to run the pipeline on. -1 for CPU (default), 0 for the first GPU and so on.",
+    )
+    args = parser.parse_args()
+
+    main(args)

full_eval.sh

@@ -0,0 +1,15 @@
+# CV 8 - TEST
+
+python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-french --dataset mozilla-foundation/common_voice_8_0 --config fr --split test --log_outputs --greedy
+mv log_mozilla-foundation_common_voice_8_0_fr_test_predictions.txt log_mozilla-foundation_common_voice_8_0_fr_test_predictions_greedy.txt
+mv mozilla-foundation_common_voice_8_0_fr_test_eval_results.txt mozilla-foundation_common_voice_8_0_fr_test_eval_results_greedy.txt
+
+python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-french --dataset mozilla-foundation/common_voice_8_0 --config fr --split test --log_outputs
+
+# HF EVENT - DEV
+
+python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-french --dataset speech-recognition-community-v2/dev_data --config fr --split validation --chunk_length_s 5.0 --stride_length_s 1.0 --log_outputs --greedy
+mv log_speech-recognition-community-v2_dev_data_fr_validation_predictions.txt log_speech-recognition-community-v2_dev_data_fr_validation_predictions_greedy.txt
+mv speech-recognition-community-v2_dev_data_fr_validation_eval_results.txt speech-recognition-community-v2_dev_data_fr_validation_eval_results_greedy.txt
+
+python eval.py --model_id jonatasgrosman/wav2vec2-xls-r-1b-french --dataset speech-recognition-community-v2/dev_data --config fr --split validation --chunk_length_s 5.0 --stride_length_s 1.0 --log_outputs

language_model/attrs.json

@@ -0,0 +1 @@
+{"alpha": 0.5, "beta": 1.5, "unk_score_offset": -10.0, "score_boundary": true}

language_model/unigrams.txt

@@ -0,0 +1,2630 @@
+(adult_support
+</s>
+<s>
+a
+achètre
+acuillère
+adul
+adult
+agil
+agilé
+aie
+aigui
+aiguière
+aimelle
+air
+ajaje
+ak
+akalada
+akapade
+akaskalade
+akasse
+akateur
+alpaakapade
+alève
+amo
+amour
+amu
+an
+ananaoute
+angalche
+angeu
+angl
+anglache
+anglage
+anglageu
+anglaggue
+anglaye
+anglaze
+anglège
+anglé
+angèle
+anjloge
+anjo
+anklage
+anour
+aou
+aoute
+août
+apapaaparitian
+aparipion
+aparitssion
+aparsion
+aparssion
+apartion
+apartipari
+apartizion
+aparéssion
+aparétion
+aper
+apotion
+appar
+apparition
+appartirssion
+apparédition
+apparéssion
+apparétion
+appozilti
+apprécian
+apprétien
+ar
+arabe
+aragil
+arargil
+arargliaguil
+archchestre
+archestre
+archil
+archéleu
+archétreu
+argile
+argril
+arguil
+arguile
+arguileu
+arguilé
+arkestre
+arkiche
+arré
+aréar
+arégil
+ascalade
+ascalope
+askalope
+asskajope
+atrifle
+azé
+aé
+aéskalopé
+b
+baandi
+babante
+babaskét
+badi
+baie
+baille
+bain
+bainbante
+bainte
+bak
+baké
+bala
+balade
+baladeu
+balai
+balansde
+balante
+balin
+balon
+balé
+ban
+banbandi
+banbantite
+bande
+bandi
+bandit
+bane
+baneur
+banfonge
+banieul
+bannajeu
+bante
+banteu
+baroce
+bask
+baske
+baskede
+basket
+baskette
+baskeuil
+baské
+basse
+bati
+baton
+batte
+baye
+bbréteu
+be
+beberné
+bebeul
+beel
+befagueu
+bei
+beli
+bemurence
+ben
+beneur
+bente
+beui
+beuil
+beureu
+beurné
+beyuil
+beèl
+bien
+biente
+bieu
+bif
+bijelle
+bijo
+bindi
+binfage
+binjo
+binte
+binteu
+biro
+biyo
+bl
+bla
+blanc
+blank
+blé
+bléfassion
+boche
+boi
+boil
+boin
+boire
+bolieu
+bolotte
+bomon
+bon
+bonbente
+bonbonde
+bonbonfage
+bonbonheur
+bonbononeur
+bonbumon
+bonde
+bonebonheur
+boneubonheur
+bonfa
+bonfach
+bonfage
+bonfajeu
+bonheur
+bonheure
+bonjageu
+bonnebunebon
+bonte
+bonteu
+bonté
+bonvage
+bonvaléskence
+borniai
+borté
+bou
+bouette
+boufage
+boufoi
+boulboulan
+boule
+bouleau
+bouleu
+boullo
+boulon
+boulot
+boulotte
+bounheur
+boyl
+boèle
+bra
+bre
+breu
+breuil
+breuté
+bronbage
+brou
+bru
+brudurence
+brufondulence
+brui
+brume
+brumedurence
+brumeu
+brumibubrumidurence
+brumiderence
+brumidi
+brumidrence
+brumidu
+brumiduren
+brumidurence
+brumidurenceu
+brumidurenze
+brumiduresen
+brumindurence
+brumitrudurement
+brumiturence
+brumudirenk
+brumudurence
+brun
+brune
+brunidurence
+brèbrète
+brème
+brète
+bré
+bu
+bubei
+bubuman
+bubumon
+buchurelle
+bucurelle
+buie
+buil
+bukerelle
+bukrelle
+bukulbukurelle
+bukurle
+bukuré
+bulement
+bulence
+bulié
+buman
+bumisse
+bumo
+bumon
+bun
+buno
+bunon
+burkuresse
+burme
+burugreille
+bururelle
+bute
+buturelle
+bz
+bè
+bète
+béil
+béjér
+bér
+béret
+bérote
+béré
+béské
+béssan
+bézir
+c
+ca
+cable
+cacatastrophe
+cacorpon
+cafté
+cahier
+cala
+calié
+camande
+camar
+cande
+canjo
+car
+cara
+caraféfelle
+caramel
+caraval
+caravaneu
+caraveleu
+caravelle
+caravéveleu
+carcardon
+carcordonne
+cardon
+careu
+carré
+carsotrofeu
+cartaffrofe
+carte
+carton
+cartre
+caré
+cascatastrof
+cascatastrophe
+casier
+caskaderof
+casscazié
+casser
+cassi
+cassier
+cassion
+cassière
+cassié
+casta
+castastrof
+castastrophe
+castatastrophe
+castatrophe
+castor
+castraf
+catastrophe
+catatof
+catisseu
+caval
+cavelle
+cazé
+ccordon
+ccouette
+celcélié
+celier
+celié
+cellier
+celécellié
+cerf
+cerfe
+cerffe
+cerfon
+ceureu
+ch
+cha
+chacaleur
+chal
+chale
+chaleur
+chalo
+chalou
+cham
+chambre
+chambreu
+chamechambre
+champre
+chan
+chanson
+chansson
+chapre
+char
+charré
+chaso
+chasschanson
+chasseu
+chasseur
+chasson
+chausson
+chazon
+chazzon
+chch
+chemichemizié
+cheminchemizier
+chemise
+chemisier
+chemizer
+chemizi
+chemizé
+chemmizer
+chenousière
+chiko
+chjoi
+cho
+choette
+choideu
+choil
+choin
+choinson
+choix
+choizi
+chou
+chouette
+chouetteu
+choute
+choutte
+chouté
+chrève
+chureu
+chègle
+chèr
+chémissié
+chémizié
+ci
+cieux
+cirque
+ciseau
+clcleffe
+clef
+clefe
+cleffe
+cleuffe
+clière
+clèfe
+clèffe
+clécléffe
+cléffe
+co
+cocommande
+coconvcoconvalézzken
+coette
+coeur
+coière
+cojo
+cokette
+colavalassize
+cole
+colère
+comandeu
+comanne
+commande
+commen
+commun
+compagne
+comun
+con
+conbalésense
+confalé
+confffa
+conje
+conjeau
+conjejo
+conjeul
+conjon
+conjé
+conmande
+connande
+connevalésence
+contisse
+contuze
+conval
+convaleksance
+convalensencente
+convalescence
+convalesens
+convalesent
+convalkslence
+convalsence
+convalékense
+convaléksen
+convaléscence
+convalésence
+convalésesence
+convalésesesense
+convaléskens
+convaléssen
+convalésssens
+convaléséan
+conzo
+cor
+corban
+cordon
+coteu
+cotrer
+cou
+coubon
+coucouette
+couette
+coul
+couleur
+couleuvre
+courdon
+courtain
+courtir
+courtié
+cousse
+coussette
+coute
+couète
+coué
+covaleco
+covalésence
+covalésonse
+covasens
+coère
+cr
+cra
+cracrâne
+cracréteu
+craie
+craine
+cran
+crane
+craneu
+cranneu
+crate
+cravalelle
+crayon
+crcrane
+cre
+creute
+creux
+crime
+crin
+crodon
+crok
+crouitien
+crte
+crâne
+crâneu
+crè
+crèfe
+crène
+crèteu
+crève
+cré
+crémè
+crémé
+crépeu
+crérèf
+créssizion
+crétan
+créteu
+créveu
+crêpe
+crête
+cucutisse
+cui
+cuisine
+cuisiner
+cuisinier
+cuisinière
+cuisiné
+cuissetisse
+cuissin
+cuissineu
+cuissinier
+cuisson
+cuizine
+cuiziné
+culeu
+cum
+curelle
+cutice
+cutir
+cutisseu
+cutissse
+cutize
+cutizeu
+cutécutigue
+cuzinié
+cygogne
+cé
+cécé
+célié
+cérf
+cérfe
+d
+da
+daba
+dabe
+dadante
+dafffré
+damadéménagé
+dane
+dante
+davan
+davdafon
+dave
+daveau
+davon
+de
+dedalvo
+dedoufaj
+demain
+deman
+demande
+demander
+demandeu
+demandé
+demindemanagemen
+demonde
+dente
+dessiner
+detonde
+deu
+dev
+devo
+devé
+deère
+di
+diche
+didisse
+dieu
+dieux
+diga
+digace
+digacelle
+digaceuleu
+digacélé
+digacéséle
+digalcèle
+digalselle
+digalèce
+digcèle
+digigo
+diglacelle
+digo
+dij
+dijacelle
+dijaslé
+dijijo
+dik
+dikacelle
+dikacette
+dikakelle
+dimanche
+din
+dinmelle
+dinoi
+dinssion
+dire
+diré
+divan
+diveu
+dix
+diè
+diège
+diègue
+dièrge
+dié
+diége
+diégeu
+diégueu
+diégé
+do
+doi
+doing
+doir
+don
+dondonteu
+dondé
+donedé
+donfage
+donfaje
+donné
+donte
+donté
+doonte
+dote
+doujoute
+doukkreukrèle
+douveu
+dproduikssion
+dralation
+drdrdrdr
+drevo
+drume
+drècrète
+drève
+dréfonette
+du
+duddubute
+dudu
+duf
+duie
+duil
+dukacelle
+dumon
+dupuè
+durlène
+duvduvé
+duve
+duverre
+duvet
+duvette
+duvo
+duvon
+dué
+dé
+débal
+débar
+débarakemen
+débarakmen
+débardemen
+débardement
+débarken
+débarkément
+débarquement
+débéret
+décalartion
+déclar
+déclaration
+déclaratsion
+déclarition
+dédémandéménagé
+défi
+déglindéklarassion
+dékla
+déklaratsion
+déklarssion
+déklassion
+dékrarassion
+délaratio
+déman
+démanageme
+démande
+démangenassionage
+démannagement
+démar
+démarken
+démen
+démengé
+démeunagemen
+démingemen
+démédéménagedéméunnadéménagemen
+déménage
+déménagemen
+déménagement
+déménagerment
+déménajmen
+déménajment
+dép
+dépa
+dépara
+dépardébarquemen
+déparen
+déparkemen
+déparquement
+départdépartemen
+département
+départeur
+départinman
+départtemen
+dérbéré
+déré
+déscala
+désir
+déss
+déssdézir
+déssi
+déssir
+désur
+détarkemen
+détromeb
+dévo
+dézi
+e
+eca
+ekuatèreteur
+ekuiaté
+emposition
+en
+enjlanej
+enkla
+enposition
+eper
+epuji
+equateur
+eruite
+es
+escala
+escalade
+escalader
+escalope
+escalé
+escapade
+escapate
+escar
+escargo
+escargot
+esclaeskalope
+esclave
+escr
+escroc
+escrok
+esescapade
+eseskapade
+eseskorte
+esgabade
+eskabbade
+eskaeskalope
+eskaladeu
+eskalalope
+eskapa
+eskape
+eskarlope
+eskaron
+eski
+eskitassion
+esklapin
+espade
+esscrok
+essitatoine
+estragon
+estrokonkon
+estrumen
+etchateur
+etra
+etsitation
+eu
+eucateur
+euh
+eukateur
+euli
+eum
+euscallope
+euskapade
+euskro
+eustrumain
+ex
+excalope
+excapade
+excitation
+excétation
+existassion
+exkro
+extinteur
+ezalope
+f
+fa
+face
+facile
+fafème
+fage
+faire
+fakade
+fakadé
+faleur
+fameu
+fari
+farine
+farkile
+fass
+fassadeu
+fassate
+fasse
+fassé
+façade
+façadeu
+façadé
+feme
+femme
+fente
+fere
+ferme
+feu
+feufème
+feumeu
+feur
+fezesoeur
+ffar
+ffflinfodireke
+ffisse
+ffossil
+ffréfrère
+ffureur
+ffuzile
+fi
+fidomelle
+fil
+file
+fille
+fils
+filsse
+findre
+fine
+finir
+finit
+finte
+fireu
+fireur
+firo
+firoie
+firoir
+firoireu
+fitivo
+flandobirence
+flanobirence
+flemme
+fleu
+flinbirence
+flinnobirance
+flinobirence
+flisse
+flobirence
+flonobarence
+flonobirence
+flumiflunobirence
+flunobience
+flunobirence
+flunobrience
+flunobrérence
+flunoburence
+flure
+fo
+foché
+fofossile
+fofossé
+foi
+foieutte
+foire
+foireu
+foireux
+foirte
+foiss
+foissil
+foleur
+folisile
+folle
+fon
+fondo
+fonile
+fonobironce
+for
+fore
+forof
+foroi
+foss
+fosse
+fossi
+fossile
+fossileu
+fossé
+fouette
+four
+fourdin
+fozer
+fozé
+frer
+freu
+freuvandulence
+frevoné
+frfoir
+fro
+froduction
+frodukssion
+froi
+frorari
+frumiturence
+frère
+fré
+fréreu
+frésonse
+frézon
+ftui
+fu
+fufuzile
+fufuzzile
+fui
+ful
+funobirence
+fur
+fure
+furefureur
+fureu
+fureur
+furfureur
+furi
+furieu
+furieur
+furor
+fus
+fusieu
+fusil
+fusile
+fusié
+fussi
+fussifussile
+fussil
+fussile
+fuz
+fuzi
+fuzil
+fuzile
+fvimineminf
+fè
+fèbe
+fèfème
+fème
+fène
+fère
+fé
+féfame
+féime
+fémeu
+féreu
+férmeu
+férmé
+févreu
+g
+ga
+gachon
+gagralassion
+gaon
+garcedrape
+gardon
+garçon
+gassiyeu
+gaz
+gaçon
+gelé
+getère
+geu
+gigolle
+gigot
+gigote
+glalaggralassion
+glalassion
+glgralassion
+glope
+glumon
+glève
+goix
+gordon
+gou
+goubrin
+goudin
+gour
+gourdeu
+gourdien
+gourdin
+gourgin
+gourédin
+gragarçon
+grala
+gralation
+gralaton
+gralission
+grallazton
+gralonssion
+gralssion
+gram
+grande
+grane
+granneu
+grave
+grefe
+greuèl
+grgarçon
+grgrègrève
+griive
+grise
+grugroudon
+grèffe
+grète
+grève
+grèveu
+gré
+gréve
+gréveu
+gu
+gua
+guad
+gue
+gugo
+gui
+guigguigo
+guigo
+guigro
+guin
+gumon
+gustisse
+gué
+gélamassion
+géréal
+haie
+hopeur
+hotel
+hotèl
+hotéleu
+hupeur
+héteur
+hôtel
+i
+ien
+imposs
+impozission
+incitation
+info
+inglacege
+inglage
+ininstrumen
+inite
+inkateur
+insstrumen
+instrugmen
+instruments
+instrutmen
+insézéztararagon
+intelgence
+inteli
+intelihens
+inteliégegence
+intelligence
+intelligente
+inteuligen
+intigence
+intrumen
+intrument
+intéintéligence
+intélience
+intéligen
+intéligenz
+intéligenze
+intéliyence
+intéllijen
+intér
+intéyi
+inéksitassion
+iol
+iopposisson
+iscapade
+ite
+iège
+ièiège
+j
+ja
+jadin
+jajalou
+jajou
+jajuin
+jalon
+jaloux
+jamon
+janvjanvien
+japin
+jarbli
+jardi
+jardin
+jari
+jarjardin
+jarkanien
+jarkassion
+jarkotié
+jauf
+je
+jef
+jejeul
+jen
+jeo
+jeu
+jeuf
+jeul
+jeur
+jffuzile
+ji
+jido
+jifcalssel
+jigo
+jigotte
+jijo
+jin
+jiroir
+jiéjièl
+jje
+jjoif
+jjol
+jjuin
+jo
+joeur
+jof
+joi
+joie
+joif
+joiyeu
+jojjeul
+jokassion
+jokation
+jokatssion
+jol
+joli
+jolé
+jonbark
+jor
+joraction
+jorakassion
+jorcation
+jordan
+jorkarssion
+jorkazion
+josse
+jote
+joue
+jouer
+joujeur
+jour
+jourdin
+joué
+joveu
+jozileu
+joé
+jtinjo
+ju
+juf
+jui
+juillette
+juin
+juit
+juite
+jujuillé
+jukurelle
+junu
+jus
+jué
+jé
+jégal
+jéjande
+jérègle
+jété
+jéziègue
+ka
+kaeskalope
+kagésson
+kakakatastrof
+kaleur
+kaltié
+kaskade
+ke
+kekoeur
+keltié
+kereu
+kesér
+keu
+keureuf
+kieu
+kili
+koi
+kojo
+komate
+koué
+krcarré
+ksulisque
+kuesseu
+kui
+kurguel
+kèrfe
+kégo
+kél
+kélier
+kélk
+kér
+l
+la
+ladon
+lait
+lalardon
+lalon
+lamalédikssion
+laman
+lamon
+lanmon
+larbo
+lardo
+lardon
+lardron
+larlarbonne
+larlardon
+laver
+lejejeur
+lemèce
+len
+lenen
+lenmon
+li
+liché
+lie
+lieux
+lije
+lijo
+liju
+likoi
+liliège
+lin
+linjon
+liojlièjeu
+lion
+lir
+lis
+lisoi
+lisoie
+lisso
+lissoi
+lissoie
+lissoir
+lissoire
+livanette
+live
+livre
+lizlizoil
+lizoille
+liède
+liège
+liègeu
+lié
+liégle
+liéliège
+llamon
+lli
+llizoi
+llojie
+llé
+lo
+lode
+lof
+logis
+logisse
+logui
+loguie
+loi
+loif
+loife
+loin
+loir
+loizlizoi
+lologie
+lon
+loro
+loti
+loveu
+lui
+luif
+lumi
+lumimèce
+luminesse
+lumize
+lumizeu
+lumièce
+lumèce
+lumèke
+lumère
+luméke
+lumémèce
+lumézeu
+luni
+lurlieume
+lège
+lèt
+léande
+léche
+légen
+légende
+légendeu
+léginde
+léguande
+léguendeu
+légènede
+légé
+léjime
+léjé
+léliège
+léliègue
+léléssir
+lémlumèce
+léssoir
+m
+ma
+mai
+mail
+maizon
+maje
+malade
+malin
+malé
+malédichian
+malédicssion
+malédiction
+malédission
+malézik
+mamamalia
+mamour
+man
+mananivéleu
+manfelle
+mani
+manifé
+manimanivelle
+manivele
+manivelle
+manon
+manvielle
+marabeu
+mardé
+marmirabèle
+matik
+matin
+me
+meil
+mel
+memalédiction
+memeuil
+mendi
+menimenizé
+menizier
+meniémeil
+mennuizé
+menouizié
+menuiser
+menuisier
+menuisié
+menuisse
+menuissier
+menuissière
+menuizer
+menuizé
+menusine
+menuzier
+mené
+meo
+meu
+meuble
+meuil
+meul
+meunusier
+mfa
+mi
+michion
+midéminagemen
+miel
+mieu
+mieuil
+mieul
+mijemeu
+mil
+mileu
+mileven
+milevin
+milfené
+milieu
+mille
+million
+milon
+milvainné
+milvan
+milvane
+milvanné
+milvavène
+milven
+milvenet
+milvenette
+milveneu
+milvenmilvenné
+milvenné
+milvenu
+milvin
+milvinvène
+milvéné
+miléneven
+mimille
+mimilvinnette
+mimiziion
+mine
+minion
+minlémalédikssion
+minoi
+minon
+mion
+mirabelle
+mirabeul
+mirablbelle
+mirable
+mirableu
+miro
+mission
+misson
+mitaple
+mive
+miveu
+miye
+miyon
+mizion
+mizon
+mié
+mmaizon
+mmalédikssion
+mmenuzié
+mmeul
+mmieul
+mmillion
+mmmeuil
+mmméil
+mmoizon
+mnaan
+moi
+moichon
+moil
+moimoisson
+moireu
+moiss
+moission
+moisson
+moizin
+moizon
+momosson
+mon
+mondre
+monmoisson
+moti
+motik
+motte
+mou
+mouelle
+mouette
+mouler
+mounon
+mouton
+moutte
+moutteu
+mu
+muette
+mule
+mulen
+mumon
+munien
+muno
+munoi
+munon
+munone
+munou
+mèive
+mé
+méil
+mék
+méla
+méladikssion
+mélitétion
+mélivelle
+mélédiction
+mémoizon
+ménuizier
+ménusier
+méss
+méta
+métal
+métik
+mézion
+mézon
+n
+na
+nadré
+nadé
+nanardé
+nar
+nardeu
+nardé
+narndé
+naré
+nave
+navet
+navette
+nif
+nife
+ninnoi
+ninoie
+ninoir
+nisoigne
+nisse
+nive
+niveau
+niveu
+nivin
+nivo
+nivon
+nièle
+nmoé
+nnomembre
+nobreu
+noi
+noir
+noix
+nombre
+nombreu
+nomi
+nomouite
+non
+norme
+nou
+nounombre
+nouvé
+nu
+nuit
+nunoi
+nué
+néta
+o
+obsession
+obsition
+ocher
+ochochèstre
+ochèstretre
+ocugère
+ocujé
+ol
+olotèl
+olé
+on
+ootel
+op
+ope
+opezission
+opission
+oposission
+oposithion
+opositive
+opossission
+opozission
+opposilien
+opposission
+opposition
+oppsopeopozission
+opssion
+orch
+orchchestre
+orchestre
+orcheur
+orchtestre
+orchère
+orchèstre
+orchètreu
+orchéteu
+orkassion
+osuk
+ote
+oteje
+otlo
+oto
+ototl
+otél
+ou
+ouaoute
+oug
+oui
+oute
+oué
+p
+pa
+pabble
+pade
+padiège
+padon
+page
+pain
+painte
+pair
+pakosso
+pamon
+pan
+panbandi
+panjo
+papansson
+papati
+papppréné
+par
+parace
+parade
+paradeu
+paraparosse
+parasse
+parate
+parcor
+parcosse
+parden
+pardon
+pare
+parerosse
+pari
+park
+parkok
+paroce
+parodeu
+parohe
+parok
+paron
+paroparoce
+paroque
+parose
+parosse
+parosso
+parparasdeu
+parrosse
+parrpa
+parte
+partérission
+paré
+pasket
+paské
+paspaspaské
+passade
+passé
+pate
+pati
+patin
+paton
+peine
+peiner
+pelle
+pen
+pendeu
+pendu
+penjo
+pensan
+pente
+penteu
+pentu
+penté
+pepompi
+per
+perdan
+perné
+petite
+peu
+peul
+peureu
+phème
+pi
+pierre
+pin
+pince
+pinceau
+pinde
+pineson
+pinja
+pinjo
+pinpinzon
+pinson
+pinssan
+pinsso
+pinte
+pinzon
+pira
+pireau
+pirre
+pisson
+pizio
+pièce
+piège
+pié
+piégeu
+piéjeu
+piéjé
+plafassion
+plaftion
+plan
+plank
+pleur
+pli
+plui
+pluli
+plumidudrence
+pléfasson
+pléfation
+plépléfakssion
+plétion
+po
+podeu
+poi
+poide
+poids
+poilleu
+poin
+poipoir
+poir
+poire
+poireu
+pokarelle
+poliil
+polo
+pom
+pommier
+pompier
+pompompié
+pomé
+pon
+ponbi
+ponceur
+ponde
+ponjeur
+ponjo
+ponssoimisse
+ponte
+popromié
+poreu
+porosse
+porte
+position
+pospil
+potoi
+pouille
+poulo
+poulon
+pousser
+pparoke
+prafoné
+pre
+prefassion
+premié
+prené
+prio
+priro
+pro
+produ
+producteur
+production
+produhion
+produi
+produiprodukssion
+produission
+produitssion
+produizé
+produkprodukssion
+produktsion
+profil
+promier
+protik
+prprékizion
+prète
+pré
+prébrètze
+précision
+prédizion
+prédé
+préfassion
+préfazion
+préné
+prépréssizion
+préreu
+prés
+préssion
+préssision
+ptil
+pu
+pudi
+pui
+pukurelle
+pul
+puli
+pului
+pâté
+pème
+pé
+péme
+péné
+péplé
+pér
+péreu
+pérmé
+pérne
+pérperné
+péyerle
+que
+quelle
+r
+ra
+radission
+raie
+raifeul
+rajir
+rajo
+rak
+raklème
+ralassion
+raperaptil
+rapide
+rapleme
+raplement
+raplème
+raplé
+raplémen
+rapléraplème
+rapèl
+rapèle
+rarari
+rari
+re
+recolésence
+reconaissant
+reconaissation
+reconession
+reconnaissan
+reconnaissance
+reconnexion
+reconésonce
+reconéssan
+redré
+regale
+regar
+regarde
+regaregarer
+regaréregré
+regran
+regre
+regret
+regrette
+regrète
+reite
+rejeu
+rejo
+rejoin
+reka
+rekal
+rekokannaissa
+rekonésene
+rekonésense
+rekonéssance
+rel
+rembage
+remplèmeu
+renbage
+rengal
+renije
+renjo
+reno
+renéssance
+rep
+repli
+repti
+reptil
+reptile
+reri
+reuf
+reuil
+reul
+reupti
+reux
+rex
+reyeu
+rf
+ri
+riri
+riss
+rix
+rkkacrane
+ro
+robac
+robage
+rochon
+rof
+roglé
+rogré
+roille
+rojo
+rombage
+rombague
+rombbbage
+romebage
+ron
+ronb
+ronba
+ronbache
+ronbageu
+ronbague
+rondage
+rongajeu
+ronge
+ronje
+ronjeau
+ronjer
+ronjeu
+ronjeur
+ronjronjo
+ronpajeu
+rorcheustre
+rore
+rorozié
+rosier
+rossier
+rossié
+roujo
+rouondin
+rozer
+rrarrari
+rrerei
+rreu
+rron
+rréégl
+ru
+ruconnaissant
+rueil
+ruelle
+rugite
+rui
+ruil
+ruit
+ruite
+rupitile
+rute
+ruyi
+rège
+règle
+ré
+récal
+récit
+récite
+récitte
+réclamation
+réconsens
+réconéssanste
+rédale
+rég
+régal
+régala
+régale
+régla
+réglamation
+régol
+régrette
+régrè
+régré
+réil
+réiu
+réjen
+rékal
+rékalassion
+rékaréklamassion
+réki
+réklam
+réklamass
+rékonéssense
+rékré
+réksi
+rémé
+répiptil
+réptileu
+réptimblje
+réregré
+rérenjjo
+réréssi
+réssite
+rété
+s
+sa
+saison
+saleur
+salon
+salope
+salor
+san
+sanson
+sasaison
+satte
+scass
+sco
+se
+secret
+sekelié
+selié
+sembreu
+sension
+sentin
+sep
+seppe
+sept
+septre
+sesalade
+sesèle
+sette
+settecr
+seukrgreu
+seup
+seurf
+seuripeul
+sgourdin
+si
+siason
+sibomelle
+side
+sidomelle
+sidomesidome
+sidomène
+sidoméleu
+sidonelle
+sienjejo
+sieuge
+sigomeleu
+siiss
+sik
+sinomelle
+siroi
+sission
+situ
+situsitile
+six
+siège
+siègeu
+siègne
+sièrge
+sié
+siégeu
+siésiège
+skalate
+skré
+skèrfe
+sleu
+slèfe
+sléfe
+so
+soer
+soeur
+soi
+sonjo
+sor
+sorbil
+sorbileu
+sorbisorpilé
+sorpile
+sorpille
+sorpilo
+sorpilé
+sorpli
+sorplip
+sotonne
+soté
+souette
+sousouette
+spoporpil
+sps
+spétte
+sse
+sselié
+ssi
+ssidomelle
+ssorpil
+ssorr
+sssleffe
+ssstyle
+sstyleu
+ssuf
+st
+sta
+stab
+stade
+stadeu
+stage
+stand
+stande
+star
+state
+stin
+stiten
+strastastade
+stress
+strudemen
+stsije
+ststylé
+sttpan
+stttyle
+stuil
+style
+styleu
+stylo
+su
+sueur
+suf
+sui
+suirfené
+suisi
+suissiné
+suite
+sujel
+suli
+supo
+support
+sureur
+suzinier
+sècre
+sècrè
+sèpt
+sé
+sécré
+sédomelle
+sékré
+sémézon
+séré
+sésette
+séssan
+sésson
+sété
+séège
+t
+ta
+tab
+taba
+tabac
+tabak
+tablak
+tableu
+tabo
+tajo
+tak
+tanbé
+tanjeau
+tanjeu
+tanjeune
+tanzieu
+tap
+tapla
+tarpardission
+task
+tassier
+tataba
+tavo
+tbavo
+tchi
+tchiètche
+tchtui
+te
+teije
+tembé
+ten
+tenkssion
+tenshion
+tension
+tenson
+tente
+tentension
+tenteu
+ter
+tetisse
+teur
+tevu
+ti
+tibuné
+tieux
+tifo
+tij
+tikre
+tilieu
+timbe
+timbre
+timbreu
+time
+timebe
+timebre
+timebuné
+timepinminé
+timepinpuné
+timepune
+timepuné
+timetimbre
+timponé
+timpunet
+timpunette
+timpur
+tin
+tinjo
+tinpulé
+tinpune
+tinpunette
+tinsio
+tinsion
+tinssion
+tinte
+tipil
+tipinpune
+tipune
+tipunette
+tirl
+tite
+titinpuné
+titivo
+titjigacèle
+tive
+tiveau
+tiveo
+tiveu
+tivévé
+tiz
+tiège
+tiègre
+tière
+tié
+tiégre
+tiégue
+toi
+toir
+toire
+toit
+toite
+ton
+tonbage
+tonbé
+tonde
+tondonté
+tondé
+tonjo
+tonjoruiyé
+tonte
+tonté
+tonvaléscence
+top
+toudé
+toulotte
+tovo
+tozion
+tr
+trabition
+tradeu
+traditien
+tradition
+traditon
+traditsien
+traditssion
+tradjission
+train
+trank
+trapi
+tre
+trefon
+trefone
+trefonet
+trefonne
+trevoné
+tribe
+trime
+trin
+trinbe
+trindission
+trié
+trobo
+trofeneu
+troi
+troir
+tronc
+tronfondulence
+tronk
+trtradission
+trtréfoné
+tru
+trufodulence
+trufon
+trufondelence
+trufondulen
+trufondulence
+trufondulenze
+trufondulsen
+trulèbe
+trulème
+trulé
+trèbe
+trèbeu
+trège
+trème
+trèmejol
+trèmeu
+trèp
+trètrème
+trève
+trèème
+tré
+trébe
+trébeu
+trébi
+trébo
+trébé
+tréchfoné
+trédission
+tréfondé
+tréfoué
+trémé
+trénte
+trépe
+trétraire
+tsepé
+tstimbre
+ttrifudulence
+ttrotrefeufoé
+tu
+tudisse
+tufondusse
+tuie
+tuil
+tuile
+tuit
+tuite
+tuji
+tujil
+tulème
+tulémeu
+tumon
+tupeur
+tupune
+tur
+turfontulen
+turleme
+turlèlme
+turlème
+turlèvme
+turte
+tut
+tute
+tutisse
+tuve
+tuèl
+té
+tébré
+tégacelle
+téjé
+téligence
+témpuné
+témé
+térpin
+téré
+u
+un
+upeur
+upr
+upré
+upér
+utéligen
+vagen
+vagin
+vakan
+val
+vale
+valeor
+valet
+valette
+valeur
+valvoleur
+valéintan
+vaçade
+ven
+vendre
+venjo
+venleur
+vent
+vente
+venteu
+ventre
+vi
+vidange
+vindeu
+vingt
+vingte
+vinte
+vite
+vitive
+vitivo
+vive
+vivo
+vo
+voileur
+voir
+voix
+voleur
+voleureu
+volo
+volé
+vor
+vot
+vou
+vvaleur
+vvente
+vvoleur
+ye
+zeu
+zeusse
+zeux
+zezeu
+zezu
+zidolmelle
+zieu
+zigo
+ziège
+zor
+zorkassion
+zouinezozzou
+zu
+zuin
+zé
+ lieu
+écal
+écapade
+écho
+éclalope
+écouteur
+écro
+écu
+écuirékouère
+écuière
+écureuil
+écurieère
+écuririe
+écuyière
+écuyère
+éeu
+églaminssion
+égro
+éguillèreu
+éguière
+égzékssitassion
+éka
+ékabade
+ékadteur
+ékajade
+ékakeur
+ékalade
+ékalajade
+ékalope
+ékapadeu
+ékart
+ékarteur
+ékateu
+ékateur
+ékaékateur
+ékaéquaékaéquateur
+éke
+ékitassion
+ékkitation
+ékosse
+ékouteur
+ékrkro
+ékro
+ékrok
+éksitass
+ékstaze
+ékuière
+ékuiécureuil
+ékukuiyère
+ékunteur
+ékurié
+ékurékuiéyère
+ékuyère
+ékuékuiyère
+élieur
+élou
+élève
+élévemen
+éléveu
+épeur
+épogneu
+épozission
+épozition
+épozizion
+éra
+éragon
+és
+ésalope
+ésalopeu
+éscalope
+éscor
+éscrotke
+ésgapade
+ésita
+éska
+éskalaéskalope
+éskaléskalade
+éskapadeu
+éske
+éskor
+éskork
+éskrok
+éskrosse
+éskrégon
+éssaescalade
+éssaééskapade
+éssitassion
+éssitation
+éssssitation
+ésstringon
+ésséyer
+éstragnon
+éstraon
+éstrengon
+éstringongon
+éséskalade
+éséskapade
+éséstragon
+étage
+étar
+étassion
+étor
+étra
+étragon
+étrajon
+étrangean
+étraon
+étrayon
+étél
+étéligen
+étér
+ététeur
+évou
+éékalopeu
+éékapate
+ééskalope
+ééteur
+ôte

mozilla-foundation_common_voice_8_0_fr_test_eval_results.txt

@@ -0,0 +1,2 @@
+WER: 0.16320660093488012
+CER: 0.04213180710602369

mozilla-foundation_common_voice_8_0_fr_test_eval_results_greedy.txt

@@ -0,0 +1,2 @@
+WER: 0.168549674179772
+CER: 0.046693488978296595

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0,
+  "processor_class": "Wav2Vec2ProcessorWithLM",
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c526b57c0fa8c775ae10130211e8e6ea900c5c3b89e10ad13651f2c35326a482
+size 3850553521

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

speech-recognition-community-v2_dev_data_fr_validation_eval_results.txt

@@ -0,0 +1,2 @@
+WER: 0.1716189447145687
+CER: 0.09386097249622839

speech-recognition-community-v2_dev_data_fr_validation_eval_results_greedy.txt

@@ -0,0 +1,2 @@
+WER: 0.22343778137943454
+CER: 0.09883563498510696

tokenizer_config.json

@@ -0,0 +1,48 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": true,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "do_lower_case": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "processor_class": "Wav2Vec2ProcessorWithLM",
+  "replace_word_delimiter_char": " ",
+  "target_lang": null,
+  "tokenizer_class": "Wav2Vec2CTCTokenizer",
+  "unk_token": "<unk>",
+  "word_delimiter_token": "|"
+}

vocab.json

@@ -0,0 +1,49 @@
+{
+  "'": 5,
+  "-": 6,
+  "</s>": 2,
+  "<pad>": 0,
+  "<s>": 1,
+  "<unk>": 3,
+  "a": 7,
+  "b": 8,
+  "c": 9,
+  "d": 10,
+  "e": 11,
+  "f": 12,
+  "g": 13,
+  "h": 14,
+  "i": 15,
+  "j": 16,
+  "k": 17,
+  "l": 18,
+  "m": 19,
+  "n": 20,
+  "o": 21,
+  "p": 22,
+  "q": 23,
+  "r": 24,
+  "s": 25,
+  "t": 26,
+  "u": 27,
+  "v": 28,
+  "w": 29,
+  "x": 30,
+  "y": 31,
+  "z": 32,
+  "|": 4,
+  "à": 33,
+  "â": 34,
+  "ã": 35,
+  "ç": 36,
+  "è": 37,
+  "é": 38,
+  "ê": 39,
+  "ë": 40,
+  "î": 41,
+  "ï": 42,
+  "ô": 43,
+  "ù": 44,
+  "û": 45,
+  "œ": 46
+}