Unstable speech

[Emil-18]

The speech randomly stops and the following error is written to the log
ERROR - external:synthDrivers.piper_neural_voices.process_speech_sequence (14:37:55.572) - piper4nvda_asyncio (18500):
Failed to execute speech task <synthDrivers.piper_neural_voices.SpeechTask object at 0x00FB5850>
Traceback (most recent call last):
File "C:\Users\e-hes\AppData\Roaming\nvda\addons\piper_neural_voices\synthDrivers\piper_neural_voices_init.py", line 132, in process_speech_sequence
await callable()
File "C:\Users\e-hes\AppData\Roaming\nvda\addons\piper_neural_voices\synthDrivers\piper_neural_voices_init.py", line 104, in call
async for wave_samples in stream:
File "C:\Users\e-hes\AppData\Roaming\nvda\addons\piper_neural_voices\synthDrivers\piper_neural_voices\tts_system.py", line 186, in synthesize
async for ret in stream:
File "C:\Users\e-hes\AppData\Roaming\nvda\addons\piper_neural_voices\synthDrivers\piper_neural_voices\grpc_client_init_.py", line 148, in speak
yield ret
File "C:\Users\e-hes\AppData\Roaming\nvda\addons\piper_neural_voices\synthDrivers\piper_neural_voices\lib\grpclib\client.py", line 573, in aexit
raise exc_val
File "C:\Users\e-hes\AppData\Roaming\nvda\addons\piper_neural_voices\synthDrivers\piper_neural_voices\lib\grpclib\client.py", line 563, in aexit
await self._maybe_finish()
File "C:\Users\e-hes\AppData\Roaming\nvda\addons\piper_neural_voices\synthDrivers\piper_neural_voices\lib\grpclib\client.py", line 533, in _maybe_finish
await self.recv_trailing_metadata()
File "C:\Users\e-hes\AppData\Roaming\nvda\addons\piper_neural_voices\synthDrivers\piper_neural_voices\lib\grpclib\client.py", line 497, in recv_trailing_metadata
self._raise_for_grpc_status(status, message, details)
File "C:\Users\e-hes\AppData\Roaming\nvda\addons\piper_neural_voices\synthDrivers\piper_neural_voices\lib\grpclib\client.py", line 356, in _raise_for_grpc_status
raise GRPCError(status, message, details)
grpclib.exceptions.GRPCError: (<Status.UNKNOWN: 2>, 'Failed to obtain results', None)

This may happen in the middle of a speech sequence. When it happens, It wil omit the text that it happens on. For example if it happens on the letter D when spelling NVDA with one of the spelling commands it will say N, V, A. If it happens at the end, it will most likely speak the next speech sequence sent to it, if not this error happens on that too.