Text-to-Speech
Transformers
PyTorch
Yoruba
speecht5
text-to-audio
Inference Endpoints
AmbroseIkpele commited on
Commit
68da0f1
1 Parent(s): 95e9e4c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +32 -0
README.md CHANGED
@@ -1,3 +1,35 @@
1
  ---
2
  license: cc-by-nc-sa-4.0
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: cc-by-nc-sa-4.0
3
+ datasets:
4
+ - openslr
5
+ - mozilla-foundation/common_voice_13_0
6
+ - Lagos-NWU_Yoruba_Speech_Corpus
7
+ language:
8
+ - yo
9
+ library_name: transformers
10
+ pipeline_tag: text-to-speech
11
  ---
12
+
13
+ ```python
14
+ # Load model directly
15
+ from transformers import SpeechT5Processor, SpeechT5ForTextToSpeech, SpeechT5HifiGan
16
+ from huggingface_hub import hf_hub_download
17
+ import torch
18
+
19
+ processor = SpeechT5Processor.from_pretrained("imhotepai/yoruba-tts")
20
+ model = SpeechT5ForTextToSpeech.from_pretrained("imhotepai/yoruba-tts")
21
+
22
+ dir_= hf_hub_download(repo_id="imhotepai/yoruba-tts", filename="speaker_embeddings.pt")
23
+ speaker_embeddings= torch.load(dir_)
24
+
25
+ text='Báwó ni'.lower()
26
+ inputs = processor(text=text, return_tensors="pt")
27
+
28
+ vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan")
29
+ speech = model.generate_speech(inputs["input_ids"], speaker_embeddings, vocoder=vocoder)
30
+
31
+ # Audio in notebook
32
+ from IPython.display import Audio
33
+
34
+ Audio(speech.numpy(), rate=16000)
35
+ ```