Spaces:
Sleeping
Sleeping
File size: 5,519 Bytes
730fe87 ac171cf 730fe87 c58cbbc 35d97c8 9202468 4385b66 bd435b3 35d97c8 31bfb8d ac171cf 4385b66 35d97c8 4385b66 69f88db 4385b66 bd435b3 69f88db 4385b66 69f88db 4385b66 69f88db 4385b66 bd435b3 9202468 4385b66 bd435b3 c58cbbc bd435b3 4385b66 730fe87 4385b66 31bfb8d ac171cf 730fe87 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 |
import asyncio
import time
from chat_pipeline import ChatPipeline
from clip_transform import CLIPTransform
from chat_service import ChatService
from dotenv import load_dotenv
from speech_service import SpeechService
from concurrent.futures import ThreadPoolExecutor
from audio_stream_processor import AudioStreamProcessor
from streaming_chat_service import StreamingChatService
def time_sentance_lenghts():
load_dotenv()
print ("Initializing Chat")
# audio_processor = AudioStreamProcessor()
user_speech_service0 = SpeechService(voice_id="Adam")
prompts = [
"hello, i am a long sentance, how are you today? Tell me about your shadow self?",
"a shorter sentance",
"Jung believed that the process of self-discovery and personal growth involves confronting and integrating the shadow self into the conscious mind.",
"By doing so, we become more self-aware and more fully actualized individuals.",
]
print ("Timing prompts\n")
for prompt in prompts:
start_time = time.time()
start_stream_time = time.time()
stream = user_speech_service0.stream(prompt)
audio = b""
for chunk in stream:
if chunk is not None:
audio += chunk
end_stream_time = time.time()
from elevenlabs import play
start_speech_time = time.time()
play(audio)
end_speech_time = time.time()
end_time = time.time()
total_time = (end_time - start_time)
stream_time = (end_stream_time - start_stream_time)
speech_time = (end_speech_time - start_speech_time)
stream_multiple = speech_time / stream_time
print(f"Stream time: {stream_time:.4f}, Acutual audio time: {speech_time:.4f}, a multiple of {stream_multiple:.2f}. for prompt: {prompt}")
print ("\nChat success")
def test_sentance_lenghts():
load_dotenv()
print ("Initializing Chat")
audio_processor = AudioStreamProcessor()
user_speech_service0 = SpeechService(voice_id="Adam")
user_speech_service1 = SpeechService(voice_id="Adam")
user_speech_service2 = SpeechService(voice_id="Adam")
user_speech_service3 = SpeechService(voice_id="Adam")
prompts = [
"hello, i am a long sentance, how are you today? Tell me about your shadow self?",
"a shorter sentance",
"Jung believed that the process of self-discovery and personal growth involves confronting and integrating the shadow self into the conscious mind.",
"By doing so, we become more self-aware and more fully actualized individuals.",
]
first = True
stream1 = user_speech_service1.stream(prompts[1])
stream0 = user_speech_service0.stream(prompts[0])
time.sleep(5)
stream2 = user_speech_service2.stream(prompts[2])
stream3 = user_speech_service3.stream(prompts[3])
audio_processor.add_audio_stream(stream0)
audio_processor.add_audio_stream(stream1)
audio_processor.add_audio_stream(stream2)
audio_processor.add_audio_stream(stream3)
audio_processor.close()
from elevenlabs import generate, play
speech0 = generate(prompts[0], voice="Adam")
speech1 = generate(prompts[1], voice="Adam")
speech2 = generate(prompts[2], voice="Adam")
speech3 = generate(prompts[3], voice="Adam")
play(speech0)
play(speech1)
play(speech2)
play(speech1)
play(speech3)
play(speech1)
# for prompt in prompts:
# stream = user_speech_service.stream(prompt)
# if first:
# first = False
# time.sleep(5)
# audio_processor.add_audio_stream(stream)
audio_processor.close()
print ("Chat success")
def run_debug_code():
load_dotenv()
# print ("Initializing CLIP templates")
# clip_transform = CLIPTransform()
# print ("CLIP success")
print ("Initializing Chat")
# chat_service = ChatService()
audio_processor = AudioStreamProcessor()
chat_service = StreamingChatService(audio_processor, voice_id="2OviOUQc1JsQRQgNkVBj") # Chales003
user_speech_service = SpeechService(voice_id="Adam")
# user_speech_service.print_voices() # if you want to see your custom voices
prompts = [
"hello, how are you today?",
"tell me about your shadow self?",
"hmm, interesting, tell me more about that.",
"wait, that is so interesting, what else?",
]
for prompt in prompts:
print ("")
print (f'prompt: "{prompt}"')
stream = user_speech_service.stream(prompt)
audio_processor.add_audio_stream(stream)
print ("")
print (f'response:')
response = chat_service.respond_to(prompt)
audio_processor.close()
print ("Chat success")
async def run_pipeline():
load_dotenv()
try:
chat_pipeline = ChatPipeline()
await chat_pipeline.start()
prompts = [
"hello, how are you today?",
"tell me about your shadow self?",
"hmm, interesting, tell me more about that.",
"wait, that is so interesting, what else?",
]
for prompt in prompts:
await chat_pipeline.enqueue(prompt)
await chat_pipeline.wait_until_all_jobs_idle()
except KeyboardInterrupt:
print("Pipeline interrupted by user")
except Exception as e:
print(f"An error occurred: {e}")
if __name__ == '__main__':
# time_sentance_lenghts()
# test_sentance_lenghts()
# run_debug_code()
asyncio.run(run_pipeline())
|