File size: 6,690 Bytes
1b5c40f
 
 
 
 
 
bfe7fc0
 
1b5c40f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f7df8d8
 
 
1b5c40f
 
 
 
 
f7df8d8
 
1b5c40f
 
 
f7df8d8
1b5c40f
f7df8d8
1b5c40f
 
f7df8d8
 
 
 
 
 
 
 
 
 
 
 
 
1b5c40f
 
 
 
 
f7df8d8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1b5c40f
 
 
 
 
 
 
 
 
 
 
a3967ea
1b5c40f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
import os
import gradio as gr
import requests
import json
from moviepy import VideoFileClip
import uuid
import time
import soundfile as sf

ELEVENLABS_API_KEY = os.environ.get("ELEVENLABS_API_KEY", None)

def extract_audio(video_path, output_format="mp3"):
    if not video_path:
        return None, "No video provided"
    
    output_path = f"extracted_audio_{uuid.uuid4().hex[:8]}.{output_format}"
    
    try:
        video = VideoFileClip(video_path)
        video.audio.write_audiofile(output_path, logger=None)
        video.close()
        return output_path, f"Audio extracted successfully"
    except Exception as e:
        return None, f"Error extracting audio: {str(e)}"

def save_transcription(transcription):
    if "error" in transcription:
        return None, transcription["error"]
    transcript_filename = f"transcription_{uuid.uuid4().hex[:8]}.txt"
    
    try:
        with open(transcript_filename, "w", encoding="utf-8") as f:
            f.write(transcription.get('text', 'No text found'))
        
        return transcript_filename, "Transcription saved as text file"
    except Exception as e:
        return None, f"Error saving transcription: {str(e)}"

def process_video_file(video_file, output_format, api_key, model_id):
    if video_file is None:
        return None, "Please upload a video file", None, "No video provided"
    
    audio_path, message = extract_audio(video_file, output_format)
    
    if audio_path and os.path.exists(audio_path):
        transcription = transcribe_audio(audio_path, api_key, model_id)
        transcript_file, transcript_message = save_transcription(transcription)
        return audio_path, message, transcript_file, transcript_message
    else:
        return None, message, None, "Audio extraction failed, cannot transcribe"

def process_video_url(video_url, output_format, api_key, model_id):
    if not video_url.strip():
        return None, "Please enter a video URL", None, "No URL provided"
    
    video_path, error = download_video_from_url(video_url)
    if error:
        return None, error, None, "Video download failed, cannot transcribe"
    
    audio_path, message = extract_audio(video_path, output_format) 
    if video_path and os.path.exists(video_path):
        try:
            os.remove(video_path)
        except:
            pass
    
    if audio_path and os.path.exists(audio_path):
        transcription = transcribe_audio(audio_path, api_key, model_id)
        transcript_file, transcript_message = save_transcription(transcription)
        return audio_path, message, transcript_file, transcript_message
    else:
        return None, message, None, "Audio extraction failed, cannot transcribe"

def transcribe_audio(audio_path, api_key, model_id="scribe_v1"):
    start_time = time.time()
    
    if not api_key:
        return {"error": "Please provide an API key"}
    
    url = "https://api.elevenlabs.io/v1/speech-to-text"
    headers = {
        "xi-api-key": api_key,
        "Content-Type": "multipart/form-data"  # Explicitly set content type
    }
    
    try:
        with open(audio_path, "rb") as f:
            files = {
                "file": (os.path.basename(audio_path), f, "audio/mpeg"),
                "model_id": (None, model_id)
            }
            response = requests.post(
                url, 
                headers=headers, 
                files=files
            )
            
            # More detailed error handling
            if response.status_code != 200:
                return {
                    "error": f"API request failed with status {response.status_code}",
                    "response_text": response.text
                }
            
            result = response.json()
    except requests.exceptions.RequestException as e:
        return {"error": f"API request failed: {str(e)}"}
    except json.JSONDecodeError:
        return {"error": "Failed to parse API response"}
    except Exception as e:
        return {"error": f"Unexpected error: {str(e)}"}
    
    end_time = time.time()
    processing_time = end_time - start_time
    
    # File size calculation
    file_size = os.path.getsize(audio_path) / (1024 * 1024)
    
    # Audio duration calculation with fallback
    try:
        # Attempt to get audio duration using soundfile
        audio_data, sample_rate = sf.read(audio_path)
        audio_duration = len(audio_data) / sample_rate
    except ImportError:
        try:
            import librosa
            audio_duration = librosa.get_duration(filename=audio_path)
        except:
            audio_duration = 0
    
    # Prepare comprehensive return dictionary
    return {
        "service": "ElevenLabs Scribe",
        "text": result.get('text', ''),
        "processing_time": processing_time,
        "file_size_mb": round(file_size, 2),
        "audio_duration": round(audio_duration, 2),
        "real_time_factor": round(processing_time / audio_duration, 2) if audio_duration > 0 else None,
        "processing_speed": round(audio_duration / processing_time, 2) if processing_time > 0 else None,
        "raw_response": result
    }
with gr.Blocks(title="Video to Audio to Transcription") as app:
    gr.Markdown("# Video => Audio => Transcription")
    
    api_key = gr.Textbox(
        placeholder="Enter your ElevenLabs API key",
        label="ElevenLabs API Key",
        type="password",
        value=ELEVENLABS_API_KEY
    )
    
    model_id = gr.Dropdown(
        choices=["scribe_v1"],
        value="scribe_v1",
        label="Transcription Model"
    )
    
    with gr.Tabs():
        with gr.TabItem("Upload Video"):
            with gr.Row():
                with gr.Column():
                    video_input = gr.Video(label="Upload Video")
                    format_choice_file = gr.Radio(["mp3", "wav"], value="mp3", label="Output Format")
                    extract_button_file = gr.Button("Extract Audio & Transcribe")
                
                with gr.Column():
                    audio_output_file = gr.Audio(label="Extracted Audio", type="filepath")
                    status_output_file = gr.Textbox(label="Audio Extraction Status")
                    transcript_file_output = gr.File(label="Transcription Text File")
                    transcript_status_output = gr.Textbox(label="Transcription Status")
            
            extract_button_file.click(
                fn=process_video_file,
                inputs=[video_input, format_choice_file, api_key, model_id],
                outputs=[audio_output_file, status_output_file, transcript_file_output, transcript_status_output]
            )

if __name__ == "__main__":
    app.launch()