未验证 提交 96d76c83 编写于 作者: L liangym 提交者: GitHub

multi-spk tts static model (#2779)

* updata readme, test=doc

* update yaml and readme, test=tts

* fix batch_size, test=tts

* update readme, test=doc

* chmod, test=tts

* add multi-spk tts static model infer on server, test=tts
上级 1e839420
...@@ -247,11 +247,13 @@ class TTSServerExecutor(TTSExecutor): ...@@ -247,11 +247,13 @@ class TTSServerExecutor(TTSExecutor):
else: else:
# multi speaker do not have static model # multi speaker do not have static model
if am_dataset in {"aishell3", "vctk"}: if am_dataset in {"aishell3", "vctk"}:
pass am_result = run_model(
self.am_predictor,
[part_phone_ids.numpy(), np.array([spk_id])])
else: else:
am_result = run_model(self.am_predictor, am_result = run_model(self.am_predictor,
[part_phone_ids.numpy()]) [part_phone_ids.numpy()])
mel = am_result[0] mel = am_result[0]
self.am_time += (time.time() - am_st) self.am_time += (time.time() - am_st)
# voc # voc
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册