113 lines
2.3 KiB
Python
113 lines
2.3 KiB
Python
|
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())
|