KyraWNY commited on
Commit
a043f80
1 Parent(s): fd442ff

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +32 -0
README.md ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ datasets:
4
+ - amaai-lab/DisfluencySpeech
5
+ language:
6
+ - en
7
+ pipeline_tag: text-to-speech
8
+ ---
9
+ # Usage
10
+
11
+ ```python
12
+ from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub
13
+ from fairseq.models.text_to_speech.hub_interface import TTSHubInterface
14
+ import IPython.display as ipd
15
+
16
+ models, cfg, task = load_model_ensemble_and_task_from_hf_hub(
17
+ "amaai-lab/DisfluencySpeech_BenchmarkA",
18
+ arg_overrides={"vocoder": "hifigan", "fp16": False, "spec-bwd-max-iter": 32}
19
+ )
20
+ model = models[0]
21
+ TTSHubInterface.update_cfg_with_data_cfg(cfg, task.data_cfg)
22
+ generator = task.build_generator(models, cfg)
23
+
24
+ text = "Well, that's really funny, isn't it? What a strange world we live in."
25
+
26
+ sample = TTSHubInterface.get_model_input(task, text)
27
+ sample['net_input']['src_tokens'] = sample['net_input']['src_tokens'].cuda()
28
+ sample['net_input']['src_lengths'] = sample['net_input']['src_lengths'].cuda()
29
+ wav, rate = TTSHubInterface.get_prediction(task, model.cuda(), generator, sample)
30
+
31
+ ipd.Audio(wav.cpu(), rate=rate)
32
+ ```