Spaces:
Build error
Build error
Commit
·
fc6fb92
1
Parent(s):
e321dc1
Update app.py
Browse files
app.py
CHANGED
@@ -1,18 +1,17 @@
|
|
1 |
import gradio as gr
|
2 |
import openai
|
3 |
-
import winsound
|
4 |
from elevenlabslib import *
|
5 |
from pydub import AudioSegment
|
6 |
from pydub.playback import play
|
7 |
import io
|
8 |
-
import
|
9 |
|
10 |
openai.api_key = "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"
|
11 |
api_key = "929b67c6e533e94018a438d70c960b60"
|
12 |
from elevenlabslib import ElevenLabsUser
|
13 |
user = ElevenLabsUser(api_key)
|
14 |
|
15 |
-
messages = ["
|
16 |
|
17 |
def transcribe(audio):
|
18 |
global messages
|
@@ -25,7 +24,7 @@ def transcribe(audio):
|
|
25 |
response = openai.Completion.create(
|
26 |
engine="text-davinci-003",
|
27 |
prompt=messages[-1],
|
28 |
-
max_tokens=
|
29 |
n=1,
|
30 |
stop=None,
|
31 |
temperature=0.5,
|
@@ -34,7 +33,7 @@ def transcribe(audio):
|
|
34 |
system_message = response["choices"][0]["text"]
|
35 |
messages.append(f"{system_message}")
|
36 |
|
37 |
-
voice = user.get_voices_by_name("
|
38 |
audio = voice.generate_audio_bytes(system_message)
|
39 |
|
40 |
audio = AudioSegment.from_file(io.BytesIO(audio), format="mp3")
|
@@ -47,10 +46,9 @@ def transcribe(audio):
|
|
47 |
|
48 |
iface = gr.Interface(
|
49 |
fn=transcribe,
|
50 |
-
inputs=gr.Audio(source="microphone", type="filepath"
|
51 |
outputs="text",
|
52 |
-
title="
|
53 |
-
description="🌟 Please ask me your question and I will respond both verbally and in text to you...",
|
54 |
)
|
55 |
|
56 |
iface.launch()
|
|
|
1 |
import gradio as gr
|
2 |
import openai
|
|
|
3 |
from elevenlabslib import *
|
4 |
from pydub import AudioSegment
|
5 |
from pydub.playback import play
|
6 |
import io
|
7 |
+
import winsound
|
8 |
|
9 |
openai.api_key = "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"
|
10 |
api_key = "929b67c6e533e94018a438d70c960b60"
|
11 |
from elevenlabslib import ElevenLabsUser
|
12 |
user = ElevenLabsUser(api_key)
|
13 |
|
14 |
+
messages = ["Respond with voice"]
|
15 |
|
16 |
def transcribe(audio):
|
17 |
global messages
|
|
|
24 |
response = openai.Completion.create(
|
25 |
engine="text-davinci-003",
|
26 |
prompt=messages[-1],
|
27 |
+
max_tokens=60,
|
28 |
n=1,
|
29 |
stop=None,
|
30 |
temperature=0.5,
|
|
|
33 |
system_message = response["choices"][0]["text"]
|
34 |
messages.append(f"{system_message}")
|
35 |
|
36 |
+
voice = user.get_voices_by_name("Bella")[0]
|
37 |
audio = voice.generate_audio_bytes(system_message)
|
38 |
|
39 |
audio = AudioSegment.from_file(io.BytesIO(audio), format="mp3")
|
|
|
46 |
|
47 |
iface = gr.Interface(
|
48 |
fn=transcribe,
|
49 |
+
inputs=gr.Audio(source="microphone", type="filepath"),
|
50 |
outputs="text",
|
51 |
+
title="Voice Assistant",
|
|
|
52 |
)
|
53 |
|
54 |
iface.launch()
|