mzboito commited on
Commit
6dbaabe
·
verified ·
1 Parent(s): e89396e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -6
README.md CHANGED
@@ -16,6 +16,7 @@ pipeline_tag: automatic-speech-recognition
16
 
17
  **This is a CTC-based Automatic Speech Recognition system for French.**
18
  This model is part of the SLU demo available here: [LINK TO THE DEMO GOES HERE]
 
19
  It is based on the [mHuBERT-147](https://huggingface.co/utter-project/mHuBERT-147) speech foundation model.
20
 
21
  * Training data: XX hours
@@ -24,20 +25,21 @@ It is based on the [mHuBERT-147](https://huggingface.co/utter-project/mHuBERT-14
24
 
25
 
26
  # Table of Contents:
27
- 1. Training Parameters
28
- 2. [ASR Model class](https://huggingface.co/naver/mHuBERT-147-ASR-fr#ASR-Model-class)
29
- 3. Running inference
30
 
31
  ## Training Parameters
32
- The training parameters are available in config.yaml.
 
33
 
34
- ## ASR Model class
35
 
36
  We use the mHubertForCTC class for our model, which is nearly identical to the existing HubertForCTC class.
37
  The key difference is that we've added a few additional hidden layers at the end of the Transformer stack, just before the lm_head.
38
  The code is available in [CTC_model.py](https://huggingface.co/naver/mHuBERT-147-ASR-fr/blob/main/CTC_model.py).
39
 
40
- ## Running inference
41
 
42
  The run_asr.py file illustrates how to load the model for inference (**load_asr_model**), and how to produce transcription for a file (**run_asr_inference**).
43
  Please follow the [requirements file](https://huggingface.co/naver/mHuBERT-147-ASR-fr/blob/main/requirements.txt) to avoid incorrect model loading.
 
16
 
17
  **This is a CTC-based Automatic Speech Recognition system for French.**
18
  This model is part of the SLU demo available here: [LINK TO THE DEMO GOES HERE]
19
+
20
  It is based on the [mHuBERT-147](https://huggingface.co/utter-project/mHuBERT-147) speech foundation model.
21
 
22
  * Training data: XX hours
 
25
 
26
 
27
  # Table of Contents:
28
+ 1. [Training Parameters](https://huggingface.co/naver/mHuBERT-147-ASR-fr#Training-Parameters)
29
+ 2. [ASR Model class](https://huggingface.co/naver/mHuBERT-147-ASR-fr#ASR-Model-Class)
30
+ 3. [Running inference](https://huggingface.co/naver/mHuBERT-147-ASR-fr#Running-Inference)
31
 
32
  ## Training Parameters
33
+ The training parameters are available in [config.yaml](https://huggingface.co/naver/mHuBERT-147-ASR-fr/blob/main/config.yaml).
34
+ We highlight the use of 0.3 for hubert.final_dropout, which we found to be very helpful in convergence. We also use fp32 training, as we found fp16 training to be unstable.
35
 
36
+ ## ASR Model Class
37
 
38
  We use the mHubertForCTC class for our model, which is nearly identical to the existing HubertForCTC class.
39
  The key difference is that we've added a few additional hidden layers at the end of the Transformer stack, just before the lm_head.
40
  The code is available in [CTC_model.py](https://huggingface.co/naver/mHuBERT-147-ASR-fr/blob/main/CTC_model.py).
41
 
42
+ ## Running Inference
43
 
44
  The run_asr.py file illustrates how to load the model for inference (**load_asr_model**), and how to produce transcription for a file (**run_asr_inference**).
45
  Please follow the [requirements file](https://huggingface.co/naver/mHuBERT-147-ASR-fr/blob/main/requirements.txt) to avoid incorrect model loading.