Finish up the basic Rigel TTS

rigel/requirements.txt

@@ -1,2 +1,3 @@
 tts==0.6
 python-dotenv
+simpleaudio

rigel/server.py

@@ -23,7 +23,12 @@ from tcp import start, synth
 
 def main():
     config = dotenv.dotenv_values(sys.argv[1])
-    asyncio.run(start(config, synth(config)))
+
+    try:
+        asyncio.run(start(config, synth(config)))
+    except KeyboardInterrupt:
+        sys.exit()
+
 
 if __name__ == "__main__":
     main()

rigel/tcp.py

@@ -14,13 +14,51 @@
 #   You should have received a copy of the GNU General Public License
 
 import asyncio
+import numpy as np
 # pylint: disable=redefined-outer-name, unused-argument
 from pathlib import Path
 
+import simpleaudio as sa
+
 from TTS.utils.manage import ModelManager
 from TTS.utils.synthesizer import Synthesizer
 
 
+def to_wav_data(wav):
+    wav_norm = np.array(wav) * (32767 / max(0.01, np.max(np.abs(wav))))
+    return wav_norm.astype(np.int16)
+
+
+def play(wav):
+    #p = pyaudio.PyAudio()
+    try:
+        # open stream
+        # stream = p.open(format=p.get_format_from_width(wav.getsampwidth()),
+        #                 channels=wav.getnchannels(),
+        #                 rate=wav.getframerate(),
+        #                 output=True)
+        # stream = p.open(format=pyaudio.paInt16,
+        #                 channels=1,
+        #                 rate=22050,
+        #                 output=True)
+
+        # stream.write(to_wav_data(wav))
+
+        # # stop stream
+        # stream.stop_stream()
+        # stream.close()
+        # close PyAudio
+        # Start audio
+        play = sa.play_buffer(to_wav_data(wav), 1, 2, 22050)
+
+        # Wait for audio playback to finish before exiting
+        play.wait_done()
+
+    finally:
+        play.stop()
+        #p.terminate()
+
+
 def synth(config):
     path = Path(__file__).parent / ".models.json"
     manager = ModelManager(path)
@@ -51,12 +89,16 @@ def synth(config):
     )
 
     async def tcp_handler(reader, writer):
-        data = await reader.readuntil(b'\0')
+        # Read till EOF
+        data = await reader.read(-1)
         message = data.decode()
-        #wav = synthesizer.tts(message, speaker_idx, "None", None)
+        wav = synthesizer.tts(message, speaker_idx, "None", None)
+
+        play(wav)
+
         print(f"Received {message!r}")
 
-        writer.write("Ok")
+        writer.write(b"Ok")
         await writer.drain()
         writer.close()