auracaster/piper
3
0
Fork 0
mirror of https://github.com/pstrueb/piper.git synced 2026-04-17 22:05:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix single speaker models speaker_id arg

Browse Source 
port https://github.com/OpenVoiceOS/ovos-tts-plugin-piper/pull/15/files

some models fail if the arguments contain "sid", this commit adds a check to skip that kwarg
This commit is contained in:
JarbasAI
2023-12-09 15:05:22 +00:00
committed by GitHub
parent 0bb4cb9daf
commit b0228cba6d
1 changed files with 18 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

37
src/python_run/piper/voice.py
View File

@@ -128,13 +128,13 @@ class PiperVoice:
noise_w=noise_w, noise_w=noise_w,
) + silence_bytes ) + silence_bytes
def synthesize_ids_to_raw( def synthesize_ids_to_raw(
self, self,
phoneme_ids: List[int], phoneme_ids: List[int],
speaker_id: Optional[int] = None, speaker_id: Optional[int] = None,
length_scale: Optional[float] = None, length_scale: Optional[float] = None,
noise_scale: Optional[float] = None, noise_scale: Optional[float] = None,
noise_w: Optional[float] = None, noise_w: Optional[float] = None,
) -> bytes: ) -> bytes:
"""Synthesize raw audio from phoneme ids.""" """Synthesize raw audio from phoneme ids."""
if length_scale is None: if length_scale is None:
@@ -153,25 +153,24 @@ class PiperVoice:
dtype=np.float32, dtype=np.float32,
) )
args = {
"input": phoneme_ids_array,
"input_lengths": phoneme_ids_lengths,
"scales": scales
}
if self.config.num_speakers <= 1:
speaker_id = None
if (self.config.num_speakers > 1) and (speaker_id is None): if (self.config.num_speakers > 1) and (speaker_id is None):
# Default speaker # Default speaker
speaker_id = 0 speaker_id = 0
sid = None
if speaker_id is not None: if speaker_id is not None:
sid = np.array([speaker_id], dtype=np.int64) sid = np.array([speaker_id], dtype=np.int64)
args["sid"] = sid
# Synthesize through Onnx # Synthesize through Onnx
audio = self.session.run( audio = self.session.run(None, args, )[0].squeeze((0, 1))
None,
{
"input": phoneme_ids_array,
"input_lengths": phoneme_ids_lengths,
"scales": scales,
"sid": sid,
},
)[0].squeeze((0, 1))
audio = audio_float_to_int16(audio.squeeze()) audio = audio_float_to_int16(audio.squeeze())
return audio.tobytes() return audio.tobytes()