Merge branch 'main' into Brawlence-main

extensions/api/requirements.txt

@@ -0,0 +1 @@
+flask_cloudflared==0.0.12

extensions/api/script.py

@@ -0,0 +1,90 @@
+from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
+from threading import Thread
+from modules import shared
+from modules.text_generation import generate_reply, encode
+import json
+
+params = {
+    'port': 5000,
+}
+
+class Handler(BaseHTTPRequestHandler):
+    def do_GET(self):
+        if self.path == '/api/v1/model':
+            self.send_response(200)
+            self.end_headers()
+            response = json.dumps({
+                'result': shared.model_name
+            })
+
+            self.wfile.write(response.encode('utf-8'))
+        else:
+            self.send_error(404)
+
+    def do_POST(self):
+        content_length = int(self.headers['Content-Length'])
+        body = json.loads(self.rfile.read(content_length).decode('utf-8'))
+
+        if self.path == '/api/v1/generate':
+            self.send_response(200)
+            self.send_header('Content-Type', 'application/json')
+            self.end_headers()
+
+            prompt = body['prompt']
+            prompt_lines = [l.strip() for l in prompt.split('\n')]
+
+            max_context = body.get('max_context_length', 2048)
+
+            while len(prompt_lines) >= 0 and len(encode('\n'.join(prompt_lines))) > max_context:
+                prompt_lines.pop(0)
+
+            prompt = '\n'.join(prompt_lines)
+
+            generator = generate_reply(
+                question = prompt, 
+                max_new_tokens = body.get('max_length', 200), 
+                do_sample=True, 
+                temperature=body.get('temperature', 0.5), 
+                top_p=body.get('top_p', 1), 
+                typical_p=body.get('typical', 1), 
+                repetition_penalty=body.get('rep_pen', 1.1), 
+                encoder_repetition_penalty=1, 
+                top_k=body.get('top_k', 0), 
+                min_length=0, 
+                no_repeat_ngram_size=0, 
+                num_beams=1, 
+                penalty_alpha=0, 
+                length_penalty=1,
+                early_stopping=False,
+            )
+
+            answer = ''
+            for a in generator:
+                answer = a[0]
+
+            response = json.dumps({
+                'results': [{
+                    'text': answer[len(prompt):]
+                }]
+            })
+            self.wfile.write(response.encode('utf-8'))
+        else:
+            self.send_error(404)
+
+
+def run_server():
+    server_addr = ('0.0.0.0' if shared.args.listen else '127.0.0.1', params['port'])
+    server = ThreadingHTTPServer(server_addr, Handler)
+    if shared.args.share: 
+        try:
+            from flask_cloudflared import  _run_cloudflared
+            public_url = _run_cloudflared(params['port'], params['port'] + 1)
+            print(f'Starting KoboldAI compatible api at {public_url}/api')
+        except ImportError:
+            print('You should install flask_cloudflared manually')
+    else:
+        print(f'Starting KoboldAI compatible api at http://{server_addr[0]}:{server_addr[1]}/api')
+    server.serve_forever()
+
+def ui():
+    Thread(target=run_server, daemon=True).start()

extensions/elevenlabs_tts/script.py

@@ -1,8 +1,8 @@
 from pathlib import Path
 
 import gradio as gr
-from elevenlabslib import *
-from elevenlabslib.helpers import *
+from elevenlabslib import ElevenLabsUser
+from elevenlabslib.helpers import save_bytes_to_path
 
 params = {
     'activate': True,

extensions/gallery/script.py

@@ -76,7 +76,7 @@ def generate_html():
     return container_html
 
 def ui():
-    with gr.Accordion("Character gallery"):
+    with gr.Accordion("Character gallery", open=False):
         update = gr.Button("Refresh")
         gallery = gr.HTML(value=generate_html())
     update.click(generate_html, [], gallery)

extensions/whisper_stt/requirements.txt

@@ -0,0 +1,4 @@
+git+https://github.com/Uberi/speech_recognition.git@010382b
+openai-whisper
+soundfile
+ffmpeg

extensions/whisper_stt/script.py

@@ -0,0 +1,54 @@
+import gradio as gr
+import speech_recognition as sr
+
+input_hijack = {
+    'state': False,
+    'value': ["", ""]
+}
+
+
+def do_stt(audio, text_state=""):
+    transcription = ""
+    r = sr.Recognizer()
+
+    # Convert to AudioData
+    audio_data = sr.AudioData(sample_rate=audio[0], frame_data=audio[1], sample_width=4)
+
+    try:
+        transcription = r.recognize_whisper(audio_data, language="english", model="base.en")
+    except sr.UnknownValueError:
+        print("Whisper could not understand audio")
+    except sr.RequestError as e:
+        print("Could not request results from Whisper", e)
+
+    input_hijack.update({"state": True, "value": [transcription, transcription]})
+
+    text_state += transcription + " "
+    return text_state, text_state
+
+
+def update_hijack(val):
+    input_hijack.update({"state": True, "value": [val, val]})
+    return val
+
+
+def auto_transcribe(audio, audio_auto, text_state=""):
+    if audio is None:
+        return "", ""
+    if audio_auto:
+        return do_stt(audio, text_state)
+    return "", ""
+
+
+def ui():
+    tr_state = gr.State(value="")
+    output_transcription = gr.Textbox(label="STT-Input",
+                                      placeholder="Speech Preview. Click \"Generate\" to send",
+                                      interactive=True)
+    output_transcription.change(fn=update_hijack, inputs=[output_transcription], outputs=[tr_state])
+    audio_auto = gr.Checkbox(label="Auto-Transcribe", value=True)
+    with gr.Row():
+        audio = gr.Audio(source="microphone")
+        audio.change(fn=auto_transcribe, inputs=[audio, audio_auto, tr_state], outputs=[output_transcription, tr_state])
+        transcribe_button = gr.Button(value="Transcribe")
+        transcribe_button.click(do_stt, inputs=[audio, tr_state], outputs=[output_transcription, tr_state])