moving forward

config.json

@@ -1,29 +1,34 @@
 {
 	"voices": [
 		{
-			"name": "Ralph",
+			"name": "ETHER",
 			"model_dir": "data/tokens+models/Emerson-Nature.txt_bs=64_ns=8000_vs=5000",
 			"tokeniser_file": "data/tokens+models/Emerson-Nature.txt_bs=64_ns=8000_vs=5000/Emerson-Nature.txt_ns=5000.tokenizer.json",
 			"temperature": "0.9"
 		},
 		{
-			"name": "Jean",
+			"name": "CHAOS",
-			"model_dir": "data/tokens+models/Lafontaine-Fables[english].txt_bs=64_ns=8000_vs=5000",
+			"model_dir": "data/tokens+models/Lafontaine-Fables[english].txt_bs=64_ns=3000_vs=5000",
-			"tokeniser_file": "data/tokens+models/Lafontaine-Fables[english].txt_bs=64_ns=8000_vs=5000/Lafontaine-Fables[english].txt_ns=5000.tokenizer.json",
+			"tokeniser_file": "data/tokens+models/Lafontaine-Fables[english].txt_bs=64_ns=3000_vs=5000/Lafontaine-Fables[english].txt_ns=5000.tokenizer.json",
 			"temperature": "1.2"
 		},
 		{
-			"name": "Blake",
+			"name": "BLAKE",
-			"model_dir": "data/tokens+models/Blake-Songs-of-Innocence-and-of-Experience.txt_bs=64_ns=8000_vs=5000",
+			"model_dir": "data/tokens+models/Blake-Songs-of-Innocence-and-of-Experience.txt_bs=64_ns=3000_vs=5000",
-			"tokeniser_file": "data/tokens+models/Blake-Songs-of-Innocence-and-of-Experience.txt_bs=64_ns=8000_vs=5000/Blake-Songs-of-Innocence-and-of-Experience.txt_ns=5000.tokenizer.json",
+			"tokeniser_file": "data/tokens+models/Blake-Songs-of-Innocence-and-of-Experience.txt_bs=64_ns=3000_vs=5000/Blake-Songs-of-Innocence-and-of-Experience.txt_ns=5000.tokenizer.json",
 			"temperature": "1.5"
 		},
 		{
-			"name": "Friedrich",
+			"name": "THEO",
 			"model_dir": "data/tokens+models/Schelling-ON-THE-RELATION-OF-THE-PLASTIC-ARTS-TO-NATURE.txt_bs=64_ns=8000_vs=5000",
 			"tokeniser_file": "data/tokens+models/Schelling-ON-THE-RELATION-OF-THE-PLASTIC-ARTS-TO-NATURE.txt_bs=64_ns=8000_vs=5000/Schelling-ON-THE-RELATION-OF-THE-PLASTIC-ARTS-TO-NATURE.txt_ns=5000.tokenizer.json",
 			"temperature": "1.5"
+		},
+		{
+			"name": "POMPOM",
+			"model_dir": "data/tokens+models/NEWNATUREPOEMS.txt_bs=64_ns=5000_vs=5000",
+			"tokeniser_file": "data/tokens+models/NEWNATUREPOEMS.txt_bs=64_ns=5000_vs=5000/NEWNATUREPOEMS.txt_ns=5000.tokenizer.json",
+			"temperature": "0.5"
 		}
-
 	]
 }

speak_broadcast.py

@@ -0,0 +1,113 @@
+import argparse, json, sys, time, random
+import utterance.voice
+import utterance.utils
+import utterance.osc
+import examine.metric
+
+UTTERANCE_LEN = 16
+broadcast = None
+
+def format_str(text: str) -> str:
+	t = utterance.utils.clean(text)
+	return utterance.utils.format(t)	
+
+def broadcast_utterance(v, utterance):	
+
+	global broadcast
+
+	text = v.name.upper() + ":\n"
+
+	broadcast.utterance(text, v.channel)
+	time.sleep(2)
+
+	frags = v.fragments(utterance)
+
+	for f in frags:
+		text += f
+		broadcast.utterance(text, v.channel)
+		time.sleep(2)
+
+	broadcast.command('clear')
+	time.sleep(2)
+
+				
+def main() -> int:
+
+	global broadcast
+
+	p = argparse.ArgumentParser()
+	p.add_argument("-c", "--config", type=str, default="voice.config.json", help="configuratin file")
+	p.add_argument("-i", "--iterations", type=int, default=10, help="number of iterations")
+	args = p.parse_args()
+
+	with open(args.config) as f:
+		conf = json.load(f)
+
+	voices = []
+	for v in conf['voices']:
+		model = v['model']
+		voice = utterance.voice.Voice(name=v["name"].upper(), model=model['model_dir'], tokenizer=model['tokeniser_file'], temp=float(model["temperature"]), lenght=UTTERANCE_LEN)
+		voice.set_channel(v['osc_channel']['root'], v['osc_channel']['utterance'])
+		voices.append(voice)
+
+	broadcast = utterance.osc.OscBroadcaster(name="osc_broadcast", host=conf['host'], port=conf['port'], command_channel=conf['command_osc_channel'])
+	broadcast.start_osc()
+
+	nbr_voices = len(voices)
+
+	state = 'c'
+
+	metric = examine.metric.Metric(model_input='data/models/doc2vec.model')
+
+	s = set(range(0, nbr_voices - 1))
+
+	rindex = random.sample(s, 1)[0]
+
+	v = voices[rindex]
+	uv = v.utter_one()
+	uv = format_str(uv)
+
+	v_vec = metric.vector(uv)
+	
+
+	while state == 'c':
+		
+		candidates = random.sample(s, 3)
+
+		results = []
+		for c in candidates:
+			if c == rindex:
+				continue
+			vc = voices[c]
+			vc_texts = vc.utter_n(n=50)
+			for t in vc_texts:
+				t = format_str(t)
+				t_vec = metric.vector(t)
+				d = examine.metric.cos_dist(v_vec, t_vec)
+				results.append([d, t, c])
+
+
+		results.sort(key=lambda t: t[0], reverse=True)
+
+
+		broadcast_utterance(v, uv)
+
+
+		# new round
+
+		top = results[0]
+		rindex = top[2]
+		v = voices[rindex]
+		uv = top[1]
+		v_vec = metric.vector(top[1])
+
+
+		# state = input("Continue? ")
+
+
+
+	broadcast.terminate_osc()
+
+
+if __name__ == '__main__':
+    sys.exit(main())

speak_metric.py

@@ -1,8 +1,11 @@
 import argparse, json, sys, time, random
 import utterance.voice
 import utterance.utils
+import utterance.osc
 import examine.metric
 
+UTTERANCE_LEN = 16
+
 def format_str(text: str) -> str:
 	t = utterance.utils.clean(text)
 	return utterance.utils.format(t)	
@@ -11,20 +14,23 @@ def format_str(text: str) -> str:
 def main() -> int:
 
 	p = argparse.ArgumentParser()
-	p.add_argument("-c", "--config", type=str, default="config.json", help="configuratin file")
+	p.add_argument("-c", "--config", type=str, default="voice.config.json", help="configuratin file")
 	p.add_argument("-i", "--iterations", type=int, default=10, help="number of iterations")
 	args = p.parse_args()
 
-	print(args)
-
 	with open(args.config) as f:
 		conf = json.load(f)
 
 	voices = []
 	for v in conf['voices']:
-		voice = utterance.voice.Voice(name=v["name"].upper(), model=v['model_dir'], tokenizer=v['tokeniser_file'], temp=float(v["temperature"]), lenght=16)
+		model = v['model']
+		voice = utterance.voice.Voice(name=v["name"].upper(), model=model['model_dir'], tokenizer=model['tokeniser_file'], temp=float(model["temperature"]), lenght=UTTERANCE_LEN)
+		voice.set_channel(v['osc_channel']['root'], v['osc_channel']['utterance'])
 		voices.append(voice)
 
+	broadcast = utterance.osc.OscBroadcaster(name="osc_broadcast", host=conf['host'], port=conf['port'], command_channel=conf['command_osc_channel'])
+	broadcast.start_osc()
+
 	nbr_voices = len(voices)
 
 	state = 'c'
@@ -94,23 +100,7 @@ def main() -> int:
 
 
 
-
+	broadcast.terminate_osc()
-	# nbr_voices = len(voices)
-	# current_voice = ""
-	# for i in range(args.iterations):
-	# 	rindex = random.randint(0, nbr_voices - 1)
-	# 	v = voices[rindex]
-	# 	if v.name != current_voice:
-	# 		print("==========")
-	# 		print(v.name + ":")
-	# 		current_voice = v.name
-	# 	t = v.utter_one()
-	# 	if t != None:
-	# 		t = utterance.utils.clean(t)
-	# 		t = utterance.utils.format(t)
-	# 		print(t)
-		
-	# 	time.sleep(4)
 
 
 if __name__ == '__main__':

speak_metric_broadcast.py

@@ -0,0 +1,159 @@
+import argparse, json, sys, time, random
+import utterance.voice
+import utterance.utils
+import examine.metric
+
+from osc4py3.as_eventloop import *
+from osc4py3 import oscbuildparse
+
+voices_broadcast = {}
+conf = {}
+
+def init_broadcast():
+
+	global conf, voices_broadcast
+
+	with open('voice.config.json') as f:
+		conf = json.load(f)
+
+	# configure voices and their channels
+	for v in conf['voices']:
+		voices_broadcast[v['name']] = v['osc_channel']['root'] + v['osc_channel']['utterance']
+
+	print(voices_broadcast)		
+
+	# Start the system.
+	osc_startup()
+
+	# Make client channels to send packets.
+	osc_udp_client(conf['host'], conf['port'], "aaaa")
+
+def broadcast_utterance(name, utterance):
+
+	global conf, voices_broadcast
+
+	sentences = utterance.split('\n')
+
+	text = name + ":"
+
+	for t in sentences:
+
+		if ',' in t:
+			phrases = t.split(',')
+			for i in range(0, len(phrases)):
+				p = phrases[i]				
+				if p != "":
+					text += " " + p
+					if i != len(phrases) - 1 :
+						text += ','
+					msg = oscbuildparse.OSCMessage(voices_broadcast[name], None, [text])
+					osc_send(msg, "aaaa")
+					osc_process()
+					time.sleep(2)
+		else:
+			if text == "":
+				text += t
+			else:
+				text += "\n" + t
+			msg = oscbuildparse.OSCMessage(voices_broadcast[name], None, [text])
+			osc_send(msg, "aaaa")
+			osc_process()
+			time.sleep(2)
+
+
+
+	msg = oscbuildparse.OSCMessage(conf['command_osc_channel'], None, ["clear"])
+	osc_send(msg, "aaaa")
+	osc_process()	
+
+
+def format_str(text: str) -> str:
+	t = utterance.utils.clean(text)
+	return utterance.utils.format(t)	
+
+				
+def main() -> int:
+
+	p = argparse.ArgumentParser()
+	p.add_argument("-c", "--config", type=str, default="config.json", help="configuratin file")
+	p.add_argument("-i", "--iterations", type=int, default=10, help="number of iterations")
+	args = p.parse_args()
+
+	print(args)
+
+	with open(args.config) as f:
+		conf = json.load(f)
+
+	voices = []
+	for v in conf['voices']:
+		voice = utterance.voice.Voice(name=v["name"].upper(), model=v['model_dir'], tokenizer=v['tokeniser_file'], temp=float(v["temperature"]), lenght=7)
+		voices.append(voice)
+
+	init_broadcast()
+
+	nbr_voices = len(voices)
+
+	state = 'c'
+
+	metric = examine.metric.Metric(model_input='data/models/doc2vec.model')
+
+	s = set(range(0, nbr_voices - 1))
+
+	# rindex = random.sample(s, 1)[0]
+	rindex = 4
+
+	v = voices[rindex]
+	uv = v.utter_one()
+	uv = format_str(uv)
+
+	v_vec = metric.vector(uv)
+	
+
+	while 1:
+		
+		candidates = random.sample(s, 2)
+
+		results = []
+		for c in candidates:
+			if c == rindex:
+				continue
+			vc = voices[c]
+			vc_texts = vc.utter_n(n=25)
+			for t in vc_texts:
+				t = format_str(t)
+				t_vec = metric.vector(t)
+				d = examine.metric.cos_dist(v_vec, t_vec)
+				results.append([d, t, c])
+
+
+		results.sort(key=lambda t: t[0], reverse=True)
+
+		# print('----------------------------')
+		# print(v.name + ":")
+		# print(uv)
+		# print('----------------------------')
+
+		broadcast_utterance(v.name, uv)
+
+		# for r in results[:2]:
+		# 	print('-->' + str(r[0]))
+		# 	print(r[1])
+		# 	print('+++++++++++++++++++++++++')
+
+
+		# new round
+
+		top = results[0]
+		rindex = top[2]
+		v = voices[rindex]
+		uv = top[1]
+		v_vec = metric.vector(top[1])
+
+
+		# state = input("Continue? ")
+
+	osc_terminate()
+
+
+if __name__ == '__main__':
+    sys.exit(main())

utterance/osc.py

@@ -0,0 +1,34 @@
+from osc4py3.as_eventloop import *
+from osc4py3 import oscbuildparse
+
+class OscBroadcaster:
+
+	def __init__(self, name: str, host: str, port: str, command_channel: str):
+		self.name = name
+		self.host = host
+		self.port = port
+		self.cmd = command_channel
+		osc_udp_client(self.host, self.port, self.name)
+
+	def utterance(self, utterance: str, channel: str):
+		msg = oscbuildparse.OSCMessage(channel, None, [utterance])
+		osc_send(msg, self.name)
+		osc_process()
+
+	def command(self, command: str):
+		msg = oscbuildparse.OSCMessage(self.cmd, None, [command])
+		osc_send(msg, self.name)
+		osc_process()
+
+	@staticmethod
+	def start_osc():
+		osc_startup()
+
+	@staticmethod
+	def terminate_osc():
+		osc_terminate()
+
+
+
+
+

utterance/utils.py

@@ -1,4 +1,4 @@
-import string
+import string, regex
 
 def clean(text: str) -> str:
 
@@ -17,3 +17,32 @@ def clean(text: str) -> str:
 def format(text: str) -> str:
 
 	return text.replace('\r\n', '\n').replace('\n\n', '\n')
+
+def fragments(utterance: str):
+	frags = []
+	sentences = utterance.splitlines()
+
+	PUNCT_RE = regex.compile(r'(\p{Punctuation})')  
+
+	for s in sentences:
+		sf = PUNCT_RE.split(s)
+		cum = ""
+		for k in sf:
+			if len(k) < 1:
+				continue
+			elif len(k) > 1:
+				cum += k
+			elif k not in string.punctuation:
+				cum += k
+			else:
+				cum += k
+				frags.append(cum)
+				cum = ""
+		cum += '\n'
+		frags.append(cum)
+
+	return frags
+
+
+
+

utterance/voice.py

@@ -1,5 +1,6 @@
 from aitextgen import aitextgen
 import utterance.utils
+import regex, string
 
 class Voice:
 
@@ -16,3 +17,17 @@ class Voice:
 
 	def utter_one(self, temp: int = None, lenght: float = None) -> str:
 		return self.utter_n(n=1, temp=temp, lenght=lenght)[0]
+
+	def set_channel(self, root: str, endpoint: str):
+		self.channel = root + endpoint
+
+	def channel(self):
+		return self.channel
+
+	def fragments(self, utt: str):
+		self.utterance = utt
+		self.utterance_fragments = utterance.utils.fragments(utt)
+		return self.utterance_fragments
+
+
+

voice.config.json

@@ -0,0 +1,107 @@
+{
+   "command_osc_channel" : "/command",
+   "host" : "127.0.0.1",
+   "port" : "1111",
+   "voices" : [
+      {
+         "background" : "0xff571111",
+         "color" : "0xffc8d75d",
+         "font" : "Times New Roman",
+         "name" : "CHAOS",
+         "osc_channel" : {
+            "background" : "/background",
+            "color" : "/font/color",
+            "root" : "/chaos",
+            "size" : "/font/size",
+            "type" : "/font/type",
+            "utterance" : "/utterance"
+         },
+         "model" :{
+            "model_dir": "data/tokens+models/Lafontaine-Fables[english].txt_bs=64_ns=3000_vs=5000",
+            "tokeniser_file": "data/tokens+models/Lafontaine-Fables[english].txt_bs=64_ns=3000_vs=5000/Lafontaine-Fables[english].txt_ns=5000.tokenizer.json",
+            "temperature": "1.2"
+         },
+         "size" : 100
+      },
+      {
+         "background" : "0xff020000",
+         "color" : "0xffebf3f1",
+         "font" : "Arial-BoldMT",
+         "name" : "ETHER",
+         "osc_channel" : {
+            "background" : "/background",
+            "color" : "/font/color",
+            "root" : "/ether",
+            "size" : "/font/size",
+            "type" : "/font/type",
+            "utterance" : "/utterance"
+         },
+         "model" :{
+            "model_dir": "data/tokens+models/Emerson-Nature.txt_bs=64_ns=8000_vs=5000",
+            "tokeniser_file": "data/tokens+models/Emerson-Nature.txt_bs=64_ns=8000_vs=5000/Emerson-Nature.txt_ns=5000.tokenizer.json",
+            "temperature": "0.9"
+         },         
+         "size" : 100
+      },
+      {
+         "background" : "0xffdcb213",
+         "color" : "0xff353a39",
+         "font" : "NotoSansVai-Regular",
+         "name" : "BLAKE",
+         "osc_channel" : {
+            "background" : "/background",
+            "color" : "/font/color",
+            "root" : "/blake",
+            "size" : "/font/size",
+            "type" : "/font/type",
+            "utterance" : "/utterance"
+         },
+         "model" :{
+            "model_dir": "data/tokens+models/Blake-Songs-of-Innocence-and-of-Experience.txt_bs=64_ns=3000_vs=5000",
+            "tokeniser_file": "data/tokens+models/Blake-Songs-of-Innocence-and-of-Experience.txt_bs=64_ns=3000_vs=5000/Blake-Songs-of-Innocence-and-of-Experience.txt_ns=5000.tokenizer.json",
+            "temperature": "1.5"
+         },                  
+         "size" : 100
+      },
+      {
+         "background" : "0xff020000",
+         "color" : "0xffebf3f1",
+         "font" : "Arial-BoldMT",
+         "name" : "POMPOM",
+         "osc_channel" : {
+            "background" : "/background",
+            "color" : "/font/color",
+            "root" : "/pompom",
+            "size" : "/font/size",
+            "type" : "/font/type",
+            "utterance" : "/utterance"
+         },
+         "model" :{
+            "model_dir": "data/tokens+models/NEWNATUREPOEMS.txt_bs=64_ns=5000_vs=5000",
+            "tokeniser_file": "data/tokens+models/NEWNATUREPOEMS.txt_bs=64_ns=5000_vs=5000/NEWNATUREPOEMS.txt_ns=5000.tokenizer.json",
+            "temperature": "0.5"
+         },                           
+         "size" : 100
+      },
+      {
+         "background" : "0xff571111",
+         "color" : "0xffc8d75d",
+         "font" : "Times New Roman",
+         "name" : "THEO",
+         "osc_channel" : {
+            "background" : "/background",
+            "color" : "/font/color",
+            "root" : "/theo",
+            "size" : "/font/size",
+            "type" : "/font/type",
+            "utterance" : "/utterance"
+         },
+         "model" :{
+            "model_dir": "data/tokens+models/Schelling-ON-THE-RELATION-OF-THE-PLASTIC-ARTS-TO-NATURE.txt_bs=64_ns=8000_vs=5000",
+            "tokeniser_file": "data/tokens+models/Schelling-ON-THE-RELATION-OF-THE-PLASTIC-ARTS-TO-NATURE.txt_bs=64_ns=8000_vs=5000/Schelling-ON-THE-RELATION-OF-THE-PLASTIC-ARTS-TO-NATURE.txt_ns=5000.tokenizer.json",
+            "temperature": "1.5"
+         },         
+         "size" : 100
+      }
+   ]
+}