File size: 10,514 Bytes
76f9cd2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
"""
Test remote GPU transcription functionality
ζ΅‹θ―•θΏœη¨‹GPUθ½¬ε½•εŠŸθƒ½
"""

import pytest
import asyncio
import os
import tempfile
from pathlib import Path
from typing import Dict, Any

from src.tools.transcription_tools import transcribe_audio_file_tool
from src.services.audio_processing_service import AudioProcessingService


class TestRemoteTranscription:
    """Test remote GPU transcription integration"""
    
    def test_transcription_tools_initialization(self):
        """Test transcription tools initialization"""
        print("\nπŸ”§ Testing transcription tools initialization...")
        
        # Test that the tool can be imported
        assert transcribe_audio_file_tool is not None
        
        print("βœ… Transcription tools initialized successfully")
    
    @pytest.mark.asyncio
    async def test_create_sample_audio_file(self, temp_dir: str):
        """Test creating a sample audio file for transcription testing"""
        print("\n🎡 Creating sample audio file for testing...")
        
        import ffmpeg
        
        # Create a short sample audio file
        sample_file = os.path.join(temp_dir, "sample_audio.mp3")
        
        try:
            # Generate a short sine wave audio for testing
            (
                ffmpeg
                .input('sine=frequency=440:duration=5', f='lavfi')
                .output(sample_file, acodec='mp3', ar=16000)
                .overwrite_output()
                .run(quiet=True)
            )
            
            assert os.path.exists(sample_file)
            assert os.path.getsize(sample_file) > 0
            
            print(f"βœ… Sample audio file created: {sample_file}")
            print(f"   File size: {os.path.getsize(sample_file)} bytes")
            
            return sample_file
            
        except Exception as e:
            print(f"❌ Failed to create sample audio file: {str(e)}")
            pytest.skip(f"Failed to create sample audio file: {str(e)}")
    
    @pytest.mark.asyncio
    async def test_remote_transcription_endpoint_connectivity(self):
        """Test connectivity to remote transcription endpoint"""
        print("\n🌐 Testing remote transcription endpoint connectivity...")
        
        import aiohttp
        import json
        
        # Read endpoint config
        try:
            with open("endpoint_config.json", "r") as f:
                endpoint_config = json.load(f)
            
            endpoint_url = endpoint_config["transcribe_audio"]
            
            async with aiohttp.ClientSession() as session:
                # Test with a simple HEAD request to check if endpoint is reachable
                async with session.head(endpoint_url, timeout=10) as response:
                    print(f"βœ… Endpoint connectivity test:")
                    print(f"   URL: {endpoint_url}")
                    print(f"   Status: {response.status}")
                    print(f"   Headers: {dict(response.headers)}")
                    
                    # We expect either 200 (OK) or 405 (Method Not Allowed) for HEAD requests
                    assert response.status in [200, 405, 404], f"Unexpected status: {response.status}"
                    
        except asyncio.TimeoutError:
            print(f"⚠️  Endpoint connectivity timeout (expected if Modal is sleeping)")
            pytest.skip("Endpoint connectivity timeout")
        except Exception as e:
            print(f"⚠️  Endpoint connectivity test failed: {str(e)}")
            print("   This might be expected if Modal endpoint is not running")
    
    @pytest.mark.asyncio
    async def test_transcription_tool_interface(self, temp_dir: str):
        """Test transcription tool interface with sample audio"""
        print("\n🎀 Testing transcription tool interface...")
        
        # Create a sample audio file first
        sample_file = await self.test_create_sample_audio_file(temp_dir)
        
        try:
            # Test the transcription tool
            result = await transcribe_audio_file_tool(
                audio_file_path=sample_file,
                model_size="base",
                language="en",
                output_format="srt",
                enable_speaker_diarization=False
            )
            
            print(f"πŸ“‹ Transcription tool result:")
            print(f"   Status: {result.get('processing_status', 'unknown')}")
            print(f"   Audio file: {result.get('audio_file', 'N/A')}")
            print(f"   Model used: {result.get('model_used', 'N/A')}")
            print(f"   Duration: {result.get('audio_duration', 0):.2f}s")
            
            if result.get("processing_status") == "success":
                print(f"   TXT file: {result.get('txt_file_path', 'N/A')}")
                print(f"   SRT file: {result.get('srt_file_path', 'N/A')}")
                print(f"   Segments: {result.get('segment_count', 0)}")
                print("βœ… Transcription tool interface test successful")
                
                # Verify output files exist
                if result.get('txt_file_path'):
                    assert os.path.exists(result['txt_file_path'])
                if result.get('srt_file_path'):
                    assert os.path.exists(result['srt_file_path'])
                    
            else:
                print(f"   Error: {result.get('error_message', 'Unknown error')}")
                print("⚠️  Transcription failed (might be due to remote endpoint)")
                
        except Exception as e:
            print(f"❌ Transcription tool test failed: {str(e)}")
            print("   This might be expected if remote endpoint is not available")
    
    @pytest.mark.asyncio
    async def test_transcription_with_speaker_diarization(self, temp_dir: str):
        """Test transcription with speaker diarization enabled"""
        print("\nπŸ‘₯ Testing transcription with speaker diarization...")
        
        # Create a sample audio file
        sample_file = await self.test_create_sample_audio_file(temp_dir)
        
        try:
            # Test transcription with speaker diarization
            result = await transcribe_audio_file_tool(
                audio_file_path=sample_file,
                model_size="base",
                language="auto",
                output_format="srt",
                enable_speaker_diarization=True
            )
            
            print(f"πŸ“‹ Speaker diarization result:")
            print(f"   Status: {result.get('processing_status', 'unknown')}")
            print(f"   Speaker diarization enabled: {result.get('speaker_diarization_enabled', False)}")
            print(f"   Global speaker count: {result.get('global_speaker_count', 0)}")
            
            if result.get("processing_status") == "success":
                speaker_summary = result.get('speaker_summary', {})
                print(f"   Speaker summary: {speaker_summary}")
                print("βœ… Speaker diarization test successful")
            else:
                print(f"   Error: {result.get('error_message', 'Unknown error')}")
                print("⚠️  Speaker diarization failed (might be due to remote endpoint or HF token)")
                
        except Exception as e:
            print(f"❌ Speaker diarization test failed: {str(e)}")
            print("   This might be expected if HF token is not configured or endpoint unavailable")
    
    @pytest.mark.asyncio
    async def test_different_transcription_models(self, temp_dir: str):
        """Test transcription with different models"""
        print("\n🧠 Testing different transcription models...")
        
        sample_file = await self.test_create_sample_audio_file(temp_dir)
        
        models_to_test = ["tiny", "base", "small"]
        
        for model in models_to_test:
            print(f"\n   Testing model: {model}")
            try:
                result = await transcribe_audio_file_tool(
                    audio_file_path=sample_file,
                    model_size=model,
                    language="auto",
                    output_format="txt",
                    enable_speaker_diarization=False
                )
                
                if result.get("processing_status") == "success":
                    print(f"   βœ… {model} model: Success")
                    print(f"      Segments: {result.get('segment_count', 0)}")
                    print(f"      Duration: {result.get('audio_duration', 0):.2f}s")
                else:
                    print(f"   ⚠️  {model} model: Failed - {result.get('error_message', 'Unknown')}")
                    
            except Exception as e:
                print(f"   ❌ {model} model: Exception - {str(e)}")
    
    @pytest.mark.asyncio
    async def test_transcription_output_formats(self, temp_dir: str):
        """Test different transcription output formats"""
        print("\nπŸ“„ Testing different output formats...")
        
        sample_file = await self.test_create_sample_audio_file(temp_dir)
        
        formats_to_test = ["txt", "srt", "json"]
        
        for format_type in formats_to_test:
            print(f"\n   Testing format: {format_type}")
            try:
                result = await transcribe_audio_file_tool(
                    audio_file_path=sample_file,
                    model_size="base",
                    language="auto",
                    output_format=format_type,
                    enable_speaker_diarization=False
                )
                
                if result.get("processing_status") == "success":
                    print(f"   βœ… {format_type} format: Success")
                    
                    # Check for format-specific outputs
                    if format_type == "txt" and result.get('txt_file_path'):
                        assert os.path.exists(result['txt_file_path'])
                    elif format_type == "srt" and result.get('srt_file_path'):
                        assert os.path.exists(result['srt_file_path'])
                        
                else:
                    print(f"   ⚠️  {format_type} format: Failed - {result.get('error_message', 'Unknown')}")
                    
            except Exception as e:
                print(f"   ❌ {format_type} format: Exception - {str(e)}")


if __name__ == "__main__":
    # Run tests with verbose output
    pytest.main([__file__, "-v", "-s"])