Commit
·
33b409c
1
Parent(s):
462721f
1B model added to tables
Browse files
README.md
CHANGED
|
@@ -167,72 +167,86 @@ The model was evaluated using the following metrics:
|
|
| 167 |
**OBS!** It should be noted that the [CoRal test dataset](https://huggingface.co/datasets/alexandrainst/coral/viewer/read_aloud/test) does not contain any conversation data, whereas the model is trained for read-aloud and conversation, but is only tested on read-aloud in the [CoRal test dataset](https://huggingface.co/datasets/CoRal-project/coral/viewer/read_aloud/test).
|
| 168 |
|
| 169 |
|
| 170 |
-
|
|
| 171 |
| :----------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | --------------------------------------------------------------------------------------: | --------------------------------------------------------------------------------------: |
|
|
|
|
| 172 |
| [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% |
|
| 173 |
-
| [CoRal-project/roest-whisper-large-v1](https://huggingface.co/CoRal-project/roest-whisper-large) | 1540M | Read-aloud | **4.3% ± 0.2%** | **10.4% ± 0.3%** |
|
| 174 |
-
| [CoRal-project/roest-wav2vec2-315M-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-
|
| 175 |
| [mhenrichsen/hviske-v2](https://huggingface.co/syvai/hviske-v2) | 1540M | Read-aloud | 4.7% ± 0.2% | 11.8% ± 0.3% |
|
| 176 |
| [openai/whisper-large-v3](https://hf.co/openai/whisper-large-v3) | 1540M | - | 11.4% ± 0.3% | 28.3% ± 0.6% |
|
| 177 |
|
| 178 |
**OBS!** Benchmark for hviske-v2 has been reevaluted and the confidence interval is larger than reported in the model card.
|
| 179 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 180 |
### Detailed evaluation across demographics on the CoRal test data
|
| 181 |
<img src="https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2/resolve/main/images/wer.png">
|
| 182 |
|
| 183 |
<img src="https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2/resolve/main/images/cer.png">
|
| 184 |
|
| 185 |
### Table WER scores in % of evaluation across demographics on the CoRal test data
|
| 186 |
-
| Category | roest-whisper-large-v1 | roest-wav2vec2-315m-v1 | roest-wav2vec2-315m-v2 |
|
| 187 |
-
|
| 188 |
-
| female | 11.5 | 18.5 | 17.7 |
|
| 189 |
-
| male | 9.4 | 15.5 | 14.9 |
|
| 190 |
-
| 0-25 | 9.0 | 14.7 | 14.0 |
|
| 191 |
-
| 25-50 | 10.1 | 16.6 | 15.8 |
|
| 192 |
-
| 50+ | 11.3 | 18.2 | 17.7 |
|
| 193 |
-
| Bornholmsk | 9.8 | 17.7 | 15.7 |
|
| 194 |
-
| Fynsk | 12.1 | 18.3 | 17.7 |
|
| 195 |
-
| Københavnsk | 5.9 | 10.2 | 10.0 |
|
| 196 |
-
| Non-native | 12.2 | 20.9 | 19.4 |
|
| 197 |
-
| Nordjysk | 4.5 | 7.7 | 7.5 |
|
| 198 |
-
| Sjællandsk | 7.6 | 12.6 | 12.7 |
|
| 199 |
-
| Sydømål | 10.0 | 14.9 | 15.3 |
|
| 200 |
-
| Sønderjysk | 17.5 | 26.0 | 25.4 |
|
| 201 |
-
| Vestjysk | 15.0 | 26.3 | 25.2 |
|
| 202 |
-
| Østjysk | 7.5 | 11.7 | 11.3 |
|
| 203 |
-
| Overall | 10.4 | 17.0 | 16.3 |
|
| 204 |
|
| 205 |
|
| 206 |
### Table CER scores in % of evaluation across demographics on the CoRal test data
|
| 207 |
-
| Category | roest-whisper-large-v1 | roest-wav2vec2-315m-v1 | roest-wav2vec2-315m-v2 |
|
| 208 |
-
|
| 209 |
-
| female | 5.1 | 7.4 | 7.2 |
|
| 210 |
-
| male | 3.6 | 5.8 | 5.7 |
|
| 211 |
-
| 0-25 | 3.4 | 5.4 | 5.3 |
|
| 212 |
-
| 25-50 | 4.0 | 6.2 | 6.0 |
|
| 213 |
-
| 50+ | 5.0 | 7.5 | 7.4 |
|
| 214 |
-
| Bornholmsk | 3.8 | 6.8 | 6.1 |
|
| 215 |
-
| Fynsk | 5.1 | 7.4 | 7.2 |
|
| 216 |
-
| Københavnsk | 1.9 | 3.3 | 3.2 |
|
| 217 |
-
| Non-native | 4.8 | 7.8 | 7.5 |
|
| 218 |
-
| Nordjysk | 1.6 | 2.6 | 2.8 |
|
| 219 |
-
| Sjællandsk | 3.0 | 4.4 | 4.5 |
|
| 220 |
-
| Sydømål | 4.1 | 6.4 | 6.4 |
|
| 221 |
-
| Sønderjysk | 8.8 | 11.9 | 11.6 |
|
| 222 |
-
| Vestjysk | 6.4 | 10.1 | 9.8 |
|
| 223 |
-
| Østjysk | 2.6 | 4.0 | 4.1 |
|
| 224 |
-
| Overall | 4.3 | 6.6 | 6.5 |
|
|
|
|
| 225 |
|
| 226 |
|
| 227 |
### Roest-wav2vec2-315M with and without language model
|
| 228 |
-
The inclusion of a post-processing language model can affect the performance significantly. The Roest-v1 and Roest-v2 models are using the same Language Model (LM). The utilized LM is the one trained and used by [
|
| 229 |
|
| 230 |
-
| Model | Number of parameters | Finetuned on data of type | Postprocessed with Language Model | [CoRal](https://huggingface.co/datasets/alexandrainst/coral/viewer/read_aloud/test) CER | [CoRal](https://huggingface.
|
| 231 |
| :-------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | --------------------------------: | --------------------------------------------------------------------------------------: | --------------------------------------------------------------------------------------: |
|
|
|
|
|
|
|
| 232 |
| [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%** |
|
| 233 |
| [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% |
|
| 234 |
-
| [
|
| 235 |
-
| [
|
| 236 |
|
| 237 |
### Detailed Roest-wav2vec2-315M with and without language model on different dialects
|
| 238 |
Here are the results of the model on different danish dialects in the test set:
|
|
@@ -257,17 +271,21 @@ Here are the results of the model on different danish dialects in the test set:
|
|
| 257 |
|
| 258 |
The model was also tested against other datasets to evaluate generalizability:
|
| 259 |
|
| 260 |
-
| | **Roest-wav2vec2-315M-v1** | | **Roest-wav2vec2-315M-v2** |
|
| 261 |
-
| ------------------------------------------------------------------------------------- |
|
| 262 |
-
| Evaluation Dataset | WER %
|
| 263 |
-
| [CoRal](https://huggingface.co/datasets/
|
| 264 |
-
| [NST-da](https://huggingface.co/datasets/alexandrainst/nst-da) | 29.7
|
| 265 |
-
| [CommonVoice17](https://huggingface.co/datasets/mozilla-foundation/common_voice_17_0) | 16.7
|
| 266 |
-
| [Fleurs-da_dk](https://huggingface.co/datasets/google/fleurs) |
|
| 267 |
-
| [Fleurs-da_dk](https://huggingface.co/datasets/google/fleurs) Normed | 16.6 | 6.3 | **15.6** | **6.1** |
|
| 268 |
|
|
|
|
| 269 |
|
| 270 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 271 |
|
| 272 |
---
|
| 273 |
|
|
@@ -291,11 +309,12 @@ We would like specifically to thank Dan Saattrup Nielsen, Alexandra Institute fo
|
|
| 291 |
|
| 292 |
## Citation
|
| 293 |
|
| 294 |
-
|
| 295 |
-
|
| 296 |
@misc{roest-wav2vec2-315m-v2,
|
| 297 |
-
author = {Marie Juhl Jørgensen, Søren Vejlgaard Holm, Martin Carsten Nielsen, Dan Saattrup Nielsen, Sif Bernstorff Lehmann, Simon Leminen Madsen
|
| 298 |
title = {Roest-wav2vec-315m-v2: A Danish state-of-the-art speech recognition model trained on varied demographics and dialects},
|
| 299 |
year = {2025},
|
| 300 |
url = {https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2},
|
| 301 |
}
|
|
|
|
|
|
|
|
|
| 167 |
**OBS!** It should be noted that the [CoRal test dataset](https://huggingface.co/datasets/alexandrainst/coral/viewer/read_aloud/test) does not contain any conversation data, whereas the model is trained for read-aloud and conversation, but is only tested on read-aloud in the [CoRal test dataset](https://huggingface.co/datasets/CoRal-project/coral/viewer/read_aloud/test).
|
| 168 |
|
| 169 |
|
| 170 |
+
|Model | Number of parameters | Finetuned on data of type | [CoRal](https://huggingface.co/datasets/alexandrainst/coral/viewer/read_aloud/test) CER | [CoRal](https://huggingface.co/datasets/alexandrainst/coral/viewer/read_aloud/test) WER |
|
| 171 |
| :----------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | --------------------------------------------------------------------------------------: | --------------------------------------------------------------------------------------: |
|
| 172 |
+
| [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% |
|
| 173 |
| [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% |
|
| 174 |
+
| [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%** |
|
| 175 |
+
| [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% |
|
| 176 |
| [mhenrichsen/hviske-v2](https://huggingface.co/syvai/hviske-v2) | 1540M | Read-aloud | 4.7% ± 0.2% | 11.8% ± 0.3% |
|
| 177 |
| [openai/whisper-large-v3](https://hf.co/openai/whisper-large-v3) | 1540M | - | 11.4% ± 0.3% | 28.3% ± 0.6% |
|
| 178 |
|
| 179 |
**OBS!** Benchmark for hviske-v2 has been reevaluted and the confidence interval is larger than reported in the model card.
|
| 180 |
|
| 181 |
+
The model was also evaluated on a tentative pre-release of the coral-v2 conversation dataset. The results are tentative as the test set only includes 5 unique speakers, of which 4 are women. The test set includes 2 speakers with 'Fynsk' dialect, 1 with 'Sønderjysk', 1 with 'Non-native' and 1 'Nordjysk'. The whisper model is performing very poorly on the test set. An explanation could be hallucinations during silence and short sentences, a known whisper issue. Furthermore, both version 1 models have not been trained on any conversation data giving the models an obvious disadvantage.
|
| 182 |
+
|
| 183 |
+
| 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 |
|
| 184 |
+
| :-------------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | ------------------------------------------------------------------------------------------------------------: | ------------------------------------------------------------------------------------------------------------: |
|
| 185 |
+
| [CoRal-project/roest-wav2vec2-1B-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-1B-v2) | 1B | Read-aloud and conversation | 23.9% | 36.7% |
|
| 186 |
+
| [CoRal-project/roest-wav2vec2-315M-v2](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2) | 315M | Read-aloud and conversation | 24.2% | 37.7% |
|
| 187 |
+
| [CoRal-project/roest-whisper-large-v1](https://huggingface.co/CoRal-project/roest-whisper-large-v1) | 1540M | Read-aloud | 138% | 121% |
|
| 188 |
+
| [CoRal-project/roest-wav2vec2-315m-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v1) | 315M | Read-aloud | 123% | 80.5% |
|
| 189 |
+
|
| 190 |
+
|
| 191 |
### Detailed evaluation across demographics on the CoRal test data
|
| 192 |
<img src="https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2/resolve/main/images/wer.png">
|
| 193 |
|
| 194 |
<img src="https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2/resolve/main/images/cer.png">
|
| 195 |
|
| 196 |
### Table WER scores in % of evaluation across demographics on the CoRal test data
|
| 197 |
+
| Category | roest-whisper-large-v1 | roest-wav2vec2-315m-v1 | roest-wav2vec2-315m-v2 | roest-wav2vec2-1B-v2 |
|
| 198 |
+
|:---:|:---:|:---:|:---:|:---:|
|
| 199 |
+
| female | 11.5 | 18.5 | 17.7 | 17.8 |
|
| 200 |
+
| male | 9.4 | 15.5 | 14.9 | 15.0 |
|
| 201 |
+
| 0-25 | 9.0 | 14.7 | 14.0 | 13.7 |
|
| 202 |
+
| 25-50 | 10.1 | 16.6 | 15.8 | 15.3 |
|
| 203 |
+
| 50+ | 11.3 | 18.2 | 17.7 | 18.5 |
|
| 204 |
+
| Bornholmsk | 9.8 | 17.7 | 15.7 | 16.4 |
|
| 205 |
+
| Fynsk | 12.1 | 18.3 | 17.7 | 16.7 |
|
| 206 |
+
| Københavnsk | 5.9 | 10.2 | 10.0 | 9.5 |
|
| 207 |
+
| Non-native | 12.2 | 20.9 | 19.4 | 19.4 |
|
| 208 |
+
| Nordjysk | 4.5 | 7.7 | 7.5 | 7.3 |
|
| 209 |
+
| Sjællandsk | 7.6 | 12.6 | 12.7 | 11.0 |
|
| 210 |
+
| Sydømål | 10.0 | 14.9 | 15.3 | 14.4 |
|
| 211 |
+
| Sønderjysk | 17.5 | 26.0 | 25.4 | 27.8 |
|
| 212 |
+
| Vestjysk | 15.0 | 26.3 | 25.2 | 26.7 |
|
| 213 |
+
| Østjysk | 7.5 | 11.7 | 11.3 | 10.8 |
|
| 214 |
+
| Overall | 10.4 | 17.0 | 16.3 | 16.4 |
|
| 215 |
|
| 216 |
|
| 217 |
### Table CER scores in % of evaluation across demographics on the CoRal test data
|
| 218 |
+
| Category | roest-whisper-large-v1 | roest-wav2vec2-315m-v1 | roest-wav2vec2-315m-v2 | roest-wav2vec2-1B-v2 |
|
| 219 |
+
|:---:|:---:|:---:|:---:|:---:|
|
| 220 |
+
| female | 5.1 | 7.4 | 7.2 | 7.3 |
|
| 221 |
+
| male | 3.6 | 5.8 | 5.7 | 5.8 |
|
| 222 |
+
| 0-25 | 3.4 | 5.4 | 5.3 | 5.1 |
|
| 223 |
+
| 25-50 | 4.0 | 6.2 | 6.0 | 5.7 |
|
| 224 |
+
| 50+ | 5.0 | 7.5 | 7.4 | 7.8 |
|
| 225 |
+
| Bornholmsk | 3.8 | 6.8 | 6.1 | 6.2 |
|
| 226 |
+
| Fynsk | 5.1 | 7.4 | 7.2 | 6.9 |
|
| 227 |
+
| Københavnsk | 1.9 | 3.3 | 3.2 | 3.0 |
|
| 228 |
+
| Non-native | 4.8 | 7.8 | 7.5 | 7.3 |
|
| 229 |
+
| Nordjysk | 1.6 | 2.6 | 2.8 | 2.6 |
|
| 230 |
+
| Sjællandsk | 3.0 | 4.4 | 4.5 | 3.9 |
|
| 231 |
+
| Sydømål | 4.1 | 6.4 | 6.4 | 6.5 |
|
| 232 |
+
| Sønderjysk | 8.8 | 11.9 | 11.6 | 12.6 |
|
| 233 |
+
| Vestjysk | 6.4 | 10.1 | 9.8 | 10.5 |
|
| 234 |
+
| Østjysk | 2.6 | 4.0 | 4.1 | 3.8 |
|
| 235 |
+
| Overall | 4.3 | 6.6 | 6.5 | 6.5 |
|
| 236 |
+
|
| 237 |
|
| 238 |
|
| 239 |
### Roest-wav2vec2-315M with and without language model
|
| 240 |
+
The inclusion of a post-processing language model can affect the performance significantly. The Roest-v1 and Roest-v2 models are using the same Language Model (LM). The utilized LM is the one trained and used by [CoRal-project/roest-wav2vec2-315m-v1](https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v1).
|
| 241 |
|
| 242 |
+
| Model | Number of parameters | Finetuned on data of type | Postprocessed with Language Model | [CoRal](https://huggingface.co/datasets/alexandrainst/coral/viewer/read_aloud/test) CER | [CoRal](https://huggingface.com/datasets/alexandrainst/coral/viewer/read_aloud/test) WER |
|
| 243 |
| :-------------------------------------------------------------------------------------------- | -------------------: | --------------------------: | --------------------------------: | --------------------------------------------------------------------------------------: | --------------------------------------------------------------------------------------: |
|
| 244 |
+
| [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%** |
|
| 245 |
+
| [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% |
|
| 246 |
| [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%** |
|
| 247 |
| [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% |
|
| 248 |
+
| [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% |
|
| 249 |
+
| [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% |
|
| 250 |
|
| 251 |
### Detailed Roest-wav2vec2-315M with and without language model on different dialects
|
| 252 |
Here are the results of the model on different danish dialects in the test set:
|
|
|
|
| 271 |
|
| 272 |
The model was also tested against other datasets to evaluate generalizability:
|
| 273 |
|
| 274 |
+
| | **Roest-whisper-large-v1** | | **Roest-wav2vec2-315M-v1** | | **Roest-wav2vec2-315M-v2** | | **Roest-wav2vec2-1B-v2** | |
|
| 275 |
+
| ------------------------------------------------------------------------------------- | -------------------------- | ------- | -------------------------- | ----- | -------------------------- | ------- | ------------------------ | ------- |
|
| 276 |
+
| Evaluation Dataset | WER % | CER % | WER % | CER % | WER % | CER % | WER % | CER % |
|
| 277 |
+
| [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** |
|
| 278 |
+
| [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** |
|
| 279 |
+
| [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 |
|
| 280 |
+
| [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** |
|
|
|
|
| 281 |
|
| 282 |
+
**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.
|
| 283 |
|
| 284 |
+
---
|
| 285 |
+
### Note on comparing whisper and wav2vec2 models
|
| 286 |
+
The Whisper models detailed in this model card exhibit significantly lower Character Error Rates (CER) and Word Error Rates (WER) compared to the Wav2Vec2 models. Whisper utilizes a transformer-based architecture with additional layers that enhance contextual understanding. In contrast, Wav2Vec2 models employ shorter context windows that focus on sound prediction. The Roest-Wav2Vec2 models incorporate a straightforward language model during post-processing, which addresses errors based on statistical language patterns. Introducing a more complex, contextual post-processing language model might enable a better comparison between these model types, which the CoRal project plans to explore in future releases.
|
| 287 |
+
|
| 288 |
+
The Roest-Whisper model excels in read-aloud data, leveraging its embedded contextual framework to achieve more robust recognition within this context. However, Wav2Vec2 models appear to generalize more effectively across various speech recognition tasks, whereas Whisper models incur higher error rates in conversational data. It’s important to note that the CoRal-v2 conversation dataset, being tentative and featuring limited speaker diversity, might influence these results.
|
| 289 |
|
| 290 |
---
|
| 291 |
|
|
|
|
| 309 |
|
| 310 |
## Citation
|
| 311 |
|
| 312 |
+
```bibtex
|
|
|
|
| 313 |
@misc{roest-wav2vec2-315m-v2,
|
| 314 |
+
author = {Marie Juhl Jørgensen, Søren Vejlgaard Holm, Martin Carsten Nielsen, Dan Saattrup Nielsen, Sif Bernstorff Lehmann, Simon Leminen Madsen and Torben Blach},
|
| 315 |
title = {Roest-wav2vec-315m-v2: A Danish state-of-the-art speech recognition model trained on varied demographics and dialects},
|
| 316 |
year = {2025},
|
| 317 |
url = {https://huggingface.co/CoRal-project/roest-wav2vec2-315m-v2},
|
| 318 |
}
|
| 319 |
+
```
|
| 320 |
+
|