Update README.md

README.md

@@ -1,3 +1,74 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+---
+
+
+# Whisper Medium ATC full
+
+This model is a fine-tuned [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on Czech and English air traffic communication recordings from Czech airport LKKU.
+
+It was created as a product of bachelor's thesis at Faculty of Information Technology Brno University of Technology.
+
+# Model description
+
+
+- **Developed by:** Veronika Nevarilova ([@xnevar00](https://huggingface.co/xnevar00)), Igor Szoke ([@iszoke](https://huggingface.co/iszoke))
+- **Shared by:** [BUT FIT](https://huggingface.co/BUT-FIT)
+- **Model type:** Whisper
+- **Languages:** Czech, English
+- **License:** [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)
+- **Finetuned from model:** [openai/whisper-medium](https://huggingface.co/openai/whisper-medium)
+
+
+# Usage
+
+```
+import torch
+from transformers import pipeline
+
+audio = "path/to/audio.xx"
+device = "cuda:0" if torch.cuda.is_available() else "cpu"
+
+transcribe = pipeline(task="automatic-speech-recognition", model="whisper_full", chunk_length_s=30, device=device)
+transcribe.model.config.forced_decoder_ids = transcribe.tokenizer.get_decoder_prompt_ids(task="transcribe", language="czech")
+print('Transcription:', transcribe(audio)["text"])
+```
+
+# Dataset
+
+Training dataset was made of ~5 hours of air traffic communication recordings. Recordings were Czech and English (80:20) and sporadically Slovak.
+
+# Output format
+
+The model was learned to transcribe every recording word by word. Transcription format of a recording is as follows:
+
+Recording: *Oscar Kilo Alpha Bravo Charlie dráha dva nula střední pro přistání volná vítr nula jedna nula stupňů pět uzlů*
+
+Transcription: `Oscar Kilo Alpha Bravo Charlie dráha dva nula střední pro přistání volná vítr nula jedna nula stupňů pět uzlů`
+
+**Note:** See also model [BUT-FIT/whisper-ATC-czech-short](https://huggingface.co/BUT-FIT/whisper-ATC-czech-short), which abbreviates callsigns and numbers.
+
+
+# Results
+
+The model reached total WER of 14,7 % on unseen Czech and English LKKU recordings. 19.6 % WER was achieved on a testset containing Czech air traffic recordings from other airports, LKPR and LKTB.
+
+WER of callsings in LKKU recordings was evaluated to be 6.2 %, while on LKPR and LKTB dataset the model reached 3.6 %.
+
+# Training hyperparameters
+
+- **learning_rate:** 3e-5
+- **per_device_train_batch_size:** 2
+- **gradient_accumulation_steps:** 8
+- **warmup_ratio:** 0.12
+- **fp16:** True
+- **gradient_checkpointing:** True
+- **evaluation_strategy:** "epoch"
+- **save_strategy:** "epoch"
+- **load_best_model_at_end:** True
+- **metric_for_best_model:** "wer"
+- **num_train_epochs:** 45
+
+# Contact
+
+For further information don't hesitate to contact Veronika Nevarilova (**[xnevar00@stud.fit.vutbr.cz](xnevar00@stud.fit.vutbr.cz)**) or Igor Szoke (**[szoke@fit.vutbr.cz](szoke@fit.vutbr.cz)**).