End of training
Browse files
README.md
CHANGED
|
@@ -19,8 +19,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
# SpeechT5 Fine-Tuned on Custom Dataset
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the Custom Speech Dataset dataset.
|
| 22 |
-
It achieves the following results on the evaluation set:
|
| 23 |
-
- Loss: 0.7091
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
@@ -51,16 +49,6 @@ The following hyperparameters were used during training:
|
|
| 51 |
- training_steps: 2000
|
| 52 |
- mixed_precision_training: Native AMP
|
| 53 |
|
| 54 |
-
### Training results
|
| 55 |
-
|
| 56 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
-
|:-------------:|:------:|:----:|:---------------:|
|
| 58 |
-
| 0.0562 | 500.0 | 500 | 0.6404 |
|
| 59 |
-
| 0.0391 | 1000.0 | 1000 | 0.6750 |
|
| 60 |
-
| 0.0357 | 1500.0 | 1500 | 0.7085 |
|
| 61 |
-
| 0.0365 | 2000.0 | 2000 | 0.7091 |
|
| 62 |
-
|
| 63 |
-
|
| 64 |
### Framework versions
|
| 65 |
|
| 66 |
- Transformers 4.49.0.dev0
|
|
|
|
| 19 |
# SpeechT5 Fine-Tuned on Custom Dataset
|
| 20 |
|
| 21 |
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the Custom Speech Dataset dataset.
|
|
|
|
|
|
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 49 |
- training_steps: 2000
|
| 50 |
- mixed_precision_training: Native AMP
|
| 51 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
### Framework versions
|
| 53 |
|
| 54 |
- Transformers 4.49.0.dev0
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 577789320
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17cb8139940fda044376037a5c59f130688e1fbd88589f2ac9ef58366609200b
|
| 3 |
size 577789320
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": false,
|
| 3 |
+
"feature_extractor_type": "SpeechT5FeatureExtractor",
|
| 4 |
+
"feature_size": 1,
|
| 5 |
+
"fmax": 7600,
|
| 6 |
+
"fmin": 80,
|
| 7 |
+
"frame_signal_scale": 1.0,
|
| 8 |
+
"hop_length": 16,
|
| 9 |
+
"mel_floor": 1e-10,
|
| 10 |
+
"num_mel_bins": 80,
|
| 11 |
+
"padding_side": "right",
|
| 12 |
+
"padding_value": 0.0,
|
| 13 |
+
"processor_class": "SpeechT5Processor",
|
| 14 |
+
"reduction_factor": 2,
|
| 15 |
+
"return_attention_mask": true,
|
| 16 |
+
"sampling_rate": 16000,
|
| 17 |
+
"win_function": "hann_window",
|
| 18 |
+
"win_length": 64
|
| 19 |
+
}
|
runs/Jan14_10-50-07_42b8497ef213/events.out.tfevents.1736851808.42b8497ef213.474.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0f70977cda4338e22b845fa60fdbb60d30ffa9cd612004476e470cb5e1163b
|
| 3 |
+
size 6709
|
runs/Jan14_10-54-03_42b8497ef213/events.out.tfevents.1736852043.42b8497ef213.474.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9fd378a73dfd50ba284aa073552019487a799ebcc9650c00a70998c4061a00bd
|
| 3 |
+
size 4184
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5560
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62d389ae47aee9a1c0f7f5e1c5ab8dc224eb972c7413dd7f9c0947abcdfd08ff
|
| 3 |
size 5560
|