Hugging Face's logo

CoRal-project
/
roest-wav2vec2-315m-v2

Automatic Speech Recognition
Safetensors
Danish
wav2vec2
Eval Results
Model card Files Community
1
MarieAlvenir commited on
Commit
30f680f
·
1 Parent(s): 466b591

Indications of (This model)

Browse files
Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -187,7 +187,7 @@ Note that the high generelization error on conversation data for models trained
187
  | Model | Number of parameters | Finetuned on data of type | [CoRal-v2::conversation](https://huggingface.co/datasets/CoRal-project/coral-v2/viewer/conversation/test) CER | [CoRal-v2::conversation](https://huggingface.co/datasets/CoRal-project/coral-v2/viewer/conversation/test) WER |
188
  | :-------------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | ------------------------------------------------------------------------------------------------------------: | ------------------------------------------------------------------------------------------------------------: |
189
  | [CoRal-project/roest-wav2vec2-1B-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-1B-v2) | 1B | Read-aloud and conversation | **23.9%** | **36.7%** |
190
- | [CoRal-project/roest-wav2vec2-315M-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2) | 315M | Read-aloud and conversation | 24.2% | 37.7% |
191
  | [CoRal-project/roest-whisper-large-v1](https://huggingface.co/CoRal-project/roest-whisper-large-v1) | 1540M | Read-aloud | 138% | 121% |
192
  | [CoRal-project/roest-wav2vec2-315m-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v1) | 315M | Read-aloud | 123% | 80.5% |
193
  | [mhenrichsen/hviske-v2](https://huggingface.co/syvai/hviske-v2) | 1540M | Read-aloud | 78.2% | 72.6% |
@@ -203,7 +203,7 @@ Note that the high generelization error on conversation data for models trained
203
  | Model | Number of parameters | Finetuned on data of type | [CoRal](https://huggingface.co/datasets/CoRal-project/coral/viewer/read_aloud/test) CER | [CoRal](https://huggingface.co/datasets/CoRal-project/coral/viewer/read_aloud/test) WER |
204
  | :-------------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | --------------------------------------------------------------------------------------: | --------------------------------------------------------------------------------------: |
205
  | [CoRal-project/roest-wav2vec2-1B-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-1B-v2) | 1B | Read-aloud and conversation | 6.5% ± 0.2% | 16.4% ± 0.4% |
206
- | [CoRal-project/roest-wav2vec2-315M-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2) | 315M | Read-aloud and conversation | 6.5% ± 0.2% | 16.3% ± 0.4% |
207
  | [CoRal-project/roest-whisper-large-v1](https://huggingface.co/CoRal-project/roest-whisper-large-v1) | 1540M | Read-aloud | **4.3% ± 0.2%** | **10.4% ± 0.3%** |
208
  | [CoRal-project/roest-wav2vec2-315M-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315M-v1) | 315M | Read-aloud | 6.6% ± 0.2% | 17.0% ± 0.4% |
209
  | [mhenrichsen/hviske-v2](https://huggingface.co/syvai/hviske-v2) | 1540M | Read-aloud | 4.7% ± 0.2% | 11.8% ± 0.3% |
@@ -282,12 +282,12 @@ Note that the high generelization error on conversation data for models trained
282
  | :-------------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | --------------------------------: | --------------------------------------------------------------------------------------: | ---------------------------------------------------------------------------------------: |
283
  | [CoRal-project/roest-wav2vec2-1B-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-1B-v2) | 1B | Read-aloud and conversation | Yes | **6.5% ± 0.2%** | **16.4% ± 0.4%** |
284
  | [CoRal-project/roest-wav2vec2-1B-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-1B-v2) | 1B | Read-aloud and conversation | No | 8.1% ± 0.2% | 23.9% ± 0.4% |
285
- | [CoRal-project/roest-wav2vec2-315M-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2) | 315M | Read-aloud and conversation | Yes | **6.5% ± 0.2%** | **16.3% ± 0.4%** |
286
- | [CoRal-project/roest-wav2vec2-315M-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2) | 315M | Read-aloud and conversation | No | 8.2% ± 0.2% | 25.1% ± 0.4% |
287
  | [CoRal-project/roest-wav2vec2-315m-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v1) | 315M | Read-aloud | Yes | 6.6% ± 0.2% | 17.0% ± 0.4% |
288
  | [CoRal-project/roest-wav2vec2-315m-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v1) | 315M | Read-aloud | No | 8.6% ± 0.2% | 26.3% ± 0.5% |
289
 
290
- Here are the results of the model on different Danish dialects in the test set:
291
 
292
  | | Røst-v1 | | Røst-v1 | | Røst-v2 | | Røst-v2 | |
293
  |-------------|---------|---------|---------|---------|---------|---------|---------|---------|
@@ -314,10 +314,10 @@ The model was also tested against other datasets to evaluate generalizability:
314
  | | **Røst-whisper-large-v1** | | **Røst-wav2vec2-315M-v1** | | **Røst-wav2vec2-315M-v2** | | **Røst-wav2vec2-1B-v2** | |
315
  | ------------------------------------------------------------------------------------- | -------------------------- | --------- | -------------------------- | --------- | -------------------------- | ----------- | ------------------------ | --------- |
316
  | **Evaluation Dataset** | **WER %** | **CER %** | **WER %** | **CER %** | **WER %** | **CER %** | **WER %** | **CER %** |
317
- | [CoRal](https://huggingface.co/datasets/CoRal-project/coral/viewer/read_aloud/test) | **10.4** | **4.3** | 17.0 | 6.6 | **16.3** | **6.5** | 16.4 | **6.5** |
318
  | [NST-da](https://huggingface.co/datasets/alexandrainst/nst-da) | 29.8 | 14.5 | 29.7 | 13.9 | 26.1 | 11.9 | **12.4** | **4.9** |
319
  | [CommonVoice17](https://huggingface.co/datasets/mozilla-foundation/common_voice_17_0) | 15.6 | 8.2 | 16.7 | 6.6 | **14.4** | **5.4** | 26.3 | 10.9 |
320
- | [Fleurs-da_dk](https://huggingface.co/datasets/google/fleurs) | **12.6** | **5.1** | 16.6 | 6.3 | 15.6 | 6.1 | **13.7** | **5.5** |
321
 
322
  **OBS!** The vocab used for training incudes numerals (0,1,2,..,9), which are translated to text in a post-processing step. If the model misses spaces the numbers are interpreted as one, which especially affects the NST score as this dataset contains many numerals.
323
 
 
187
  | Model | Number of parameters | Finetuned on data of type | [CoRal-v2::conversation](https://huggingface.co/datasets/CoRal-project/coral-v2/viewer/conversation/test) CER | [CoRal-v2::conversation](https://huggingface.co/datasets/CoRal-project/coral-v2/viewer/conversation/test) WER |
188
  | :-------------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | ------------------------------------------------------------------------------------------------------------: | ------------------------------------------------------------------------------------------------------------: |
189
  | [CoRal-project/roest-wav2vec2-1B-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-1B-v2) | 1B | Read-aloud and conversation | **23.9%** | **36.7%** |
190
+ | CoRal-project/roest-wav2vec2-315M-v2 (This model)| 315M | Read-aloud and conversation | 24.2% | 37.7% |
191
  | [CoRal-project/roest-whisper-large-v1](https://huggingface.co/CoRal-project/roest-whisper-large-v1) | 1540M | Read-aloud | 138% | 121% |
192
  | [CoRal-project/roest-wav2vec2-315m-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v1) | 315M | Read-aloud | 123% | 80.5% |
193
  | [mhenrichsen/hviske-v2](https://huggingface.co/syvai/hviske-v2) | 1540M | Read-aloud | 78.2% | 72.6% |
 
203
  | Model | Number of parameters | Finetuned on data of type | [CoRal](https://huggingface.co/datasets/CoRal-project/coral/viewer/read_aloud/test) CER | [CoRal](https://huggingface.co/datasets/CoRal-project/coral/viewer/read_aloud/test) WER |
204
  | :-------------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | --------------------------------------------------------------------------------------: | --------------------------------------------------------------------------------------: |
205
  | [CoRal-project/roest-wav2vec2-1B-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-1B-v2) | 1B | Read-aloud and conversation | 6.5% ± 0.2% | 16.4% ± 0.4% |
206
+ | CoRal-project/roest-wav2vec2-315M-v2 (This model) | 315M | Read-aloud and conversation | 6.5% ± 0.2% | 16.3% ± 0.4% |
207
  | [CoRal-project/roest-whisper-large-v1](https://huggingface.co/CoRal-project/roest-whisper-large-v1) | 1540M | Read-aloud | **4.3% ± 0.2%** | **10.4% ± 0.3%** |
208
  | [CoRal-project/roest-wav2vec2-315M-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315M-v1) | 315M | Read-aloud | 6.6% ± 0.2% | 17.0% ± 0.4% |
209
  | [mhenrichsen/hviske-v2](https://huggingface.co/syvai/hviske-v2) | 1540M | Read-aloud | 4.7% ± 0.2% | 11.8% ± 0.3% |
 
282
  | :-------------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | --------------------------------: | --------------------------------------------------------------------------------------: | ---------------------------------------------------------------------------------------: |
283
  | [CoRal-project/roest-wav2vec2-1B-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-1B-v2) | 1B | Read-aloud and conversation | Yes | **6.5% ± 0.2%** | **16.4% ± 0.4%** |
284
  | [CoRal-project/roest-wav2vec2-1B-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-1B-v2) | 1B | Read-aloud and conversation | No | 8.1% ± 0.2% | 23.9% ± 0.4% |
285
+ | CoRal-project/roest-wav2vec2-315M-v2 (This model) | 315M | Read-aloud and conversation | Yes | **6.5% ± 0.2%** | **16.3% ± 0.4%** |
286
+ | CoRal-project/roest-wav2vec2-315M-v2 (This model) | 315M | Read-aloud and conversation | No | 8.2% ± 0.2% | 25.1% ± 0.4% |
287
  | [CoRal-project/roest-wav2vec2-315m-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v1) | 315M | Read-aloud | Yes | 6.6% ± 0.2% | 17.0% ± 0.4% |
288
  | [CoRal-project/roest-wav2vec2-315m-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v1) | 315M | Read-aloud | No | 8.6% ± 0.2% | 26.3% ± 0.5% |
289
 
290
+ Here are the results of the Røst-Wav2Vec2-315m models on different Danish dialects in the test set:
291
 
292
  | | Røst-v1 | | Røst-v1 | | Røst-v2 | | Røst-v2 | |
293
  |-------------|---------|---------|---------|---------|---------|---------|---------|---------|
 
314
  | | **Røst-whisper-large-v1** | | **Røst-wav2vec2-315M-v1** | | **Røst-wav2vec2-315M-v2** | | **Røst-wav2vec2-1B-v2** | |
315
  | ------------------------------------------------------------------------------------- | -------------------------- | --------- | -------------------------- | --------- | -------------------------- | ----------- | ------------------------ | --------- |
316
  | **Evaluation Dataset** | **WER %** | **CER %** | **WER %** | **CER %** | **WER %** | **CER %** | **WER %** | **CER %** |
317
+ | [CoRal](https://huggingface.co/datasets/CoRal-project/coral/viewer/read_aloud/test) | **10.4** | **4.3** | 17.0 | 6.6 | 16.3 | 6.5 | 16.4 | 6.5 |
318
  | [NST-da](https://huggingface.co/datasets/alexandrainst/nst-da) | 29.8 | 14.5 | 29.7 | 13.9 | 26.1 | 11.9 | **12.4** | **4.9** |
319
  | [CommonVoice17](https://huggingface.co/datasets/mozilla-foundation/common_voice_17_0) | 15.6 | 8.2 | 16.7 | 6.6 | **14.4** | **5.4** | 26.3 | 10.9 |
320
+ | [Fleurs-da_dk](https://huggingface.co/datasets/google/fleurs) | **12.6** | **5.1** | 16.6 | 6.3 | 15.6 | 6.1 | 13.7 | 5.5 |
321
 
322
  **OBS!** The vocab used for training incudes numerals (0,1,2,..,9), which are translated to text in a post-processing step. If the model misses spaces the numbers are interpreted as one, which especially affects the NST score as this dataset contains many numerals.
323