Merge pull request #21 from thockx/main

WIP: Extend examples

Author: kvz

examples/file_to_tts.py

@@ -0,0 +1,44 @@
+### A simple Transloadit Assembly that converts a .doc file to a .txt file. It then generates a .mp3 based on the .txt file using a text-to-speech Robot.
+
+'''
+Template:
+
+{
+  "steps": {
+    ":original": {
+      "robot": "/upload/handle"
+    },
+    "convert": {
+      "use": ":original",
+      "robot": "/document/convert",
+      "format": "txt"
+    },
+    "speech": {
+      "use": "convert",
+      "robot": "/text/speak",
+      "result": true,
+      "provider": "gcp",
+      "target_language": "en-US",
+      "voice": "female-1"
+    }
+  }
+}
+'''
+
+from transloadit import client
+
+tl = client.Transloadit('TRANSLOADIT_KEY', 'TRANSLOADIT_SECRET')
+
+def useTemplate(templateID, file_path='', result_name='', get_url=True, fields=''):
+    assembly = tl.new_assembly({'template_id': templateID, 'fields': fields})
+
+    if file_path != '':
+        assembly.add_file(open(file_path, 'rb'))
+
+    assembly_response = assembly.create(retries=5, wait=True)
+    if get_url:
+        assembly_url = assembly_response.data.get('results').get(result_name)[0].get('ssl_url')
+        print(assembly_url)
+        return assembly_url
+    
+useTemplate ('TEMPLATE_ID', file_path='fixtures/document.doc', result_name='speech', get_url=True)

examples/fixtures/crab.mp4

@@ -1,3 +1,5 @@
+### A simple Transloadit assembly that changes the size of an image.
+
 from transloadit.client import Transloadit
 
 client = Transloadit("TRANSLOADIT_KEY", "TRANSLOADIT_SECRET")

examples/fixtures/document.doc

@@ -0,0 +1,168 @@
+### A Transloadit Assembly that adds a translated TTS-voice (english to dutch) to an input video of hermit crabs.
+
+'''
+Template 1:
+
+{
+  "steps": {
+    "transcribe_json": {
+      "use": ":original",
+      "robot": "/speech/transcribe",
+      "provider": "aws",
+      "source_language": "${fields.language}",
+      "format": "json",
+      "result": true
+    }
+  }
+}
+'''
+
+'''
+Template 2:
+
+{
+  "steps": {
+    ":original": {
+      "robot": "/upload/handle"
+    },
+    "import_video": {
+      "robot": "/http/import",
+      "url": "${fields.video}"
+    },
+    "translate": {
+      "use": ":original",
+      "robot": "/text/translate",
+      "provider": "gcp",
+      "target_language": "${fields.target_language}",
+      "source_language": "${fields.source_language}",
+      "result": true
+    },
+    "speech": {
+      "use": "translate",
+      "robot": "/text/speak",
+      "provider": "gcp",
+      "target_language": "${fields.target_language}",
+      "voice": "female-1",
+      "ssml": true,
+      "result": true
+    },
+    "extract_audio": {
+      "use": "import_video",
+      "robot": "/video/encode",
+      "result": true,
+      "preset": "mp3",
+      "ffmpeg": {
+        "af": "${fields.ffmpeg}"
+      },
+      "ffmpeg_stack": "v3.3.3"
+    },
+    "merged_audio": {
+      "robot": "/audio/merge",
+      "preset": "mp3",
+      "result": "true",
+      "ffmpeg_stack": "v4.3.1",
+      "use": {
+        "steps": [
+          {
+            "name": "extract_audio",
+            "as": "audio"
+          },
+          {
+            "name": "speech",
+            "as": "audio"
+          }
+        ],
+        "volume": "sum",
+        "bundle_steps": true
+      }
+    },
+    "merged_video": {
+      "robot": "/video/merge",
+      "preset": "hls-720p",
+      "ffmpeg_stack": "v4.3.1",
+      "use": {
+        "steps": [
+          {
+            "name": "merged_audio",
+            "as": "audio"
+          },
+          {
+            "name": "import_video",
+            "as": "video"
+          }
+        ],
+        "bundle_steps": true
+      }
+    }
+  }
+}
+'''
+
+from transloadit import client
+import urllib.request
+import json
+
+tl = client.Transloadit('TRANSLOADIT_KEY', 'TRANSLOADIT_SECRET')
+
+source_language = 'en-GB'
+target_language = 'nl-NL'
+
+def useTemplate(templateID, file_path='', result_name='', get_url=True, fields=''):
+    assembly = tl.new_assembly({'template_id': templateID, 'fields': fields})
+
+    if file_path != '':
+        assembly.add_file(open(file_path, 'rb'))
+
+    assembly_response = assembly.create(retries=5, wait=True)
+    if get_url:
+        result_url = assembly_response.data.get('results').get(result_name)[0].get('ssl_url')
+        print(result_url)
+        return result_url
+    else:
+        return assembly_response
+    
+response = useTemplate ('TEMPLATE_1_ID', file_path='medium_crab.mp4', get_url=False, fields={"language":source_language})
+transcription_result_url = response.data.get('results').get('transcribe_json')[0].get('ssl_url')
+video_url = response.data.get('uploads')[0].get('ssl_url')
+
+urllib.request.urlretrieve(transcription_result_url, 'transcribe_json')
+
+with open('transcribe_json') as f:
+    data = json.load(f)
+
+ffmpeg = "volume=enable:volume=1"
+
+startTimes = []
+endTimes = []
+sentences = []
+currentSentence = ''
+
+startTimes.append(data['words'][0]['startTime'])
+
+for x in range(len(data['words'])):  
+    if (data['words'][x]['text'] == '.') and (x != len(data['words']) - 1):
+        time = data['words'][x+1]['startTime']
+        startTimes.append(time)
+    if (data['words'][x]['text'] != '.'):
+        currentSentence = currentSentence + ' ' + data['words'][x]['text']
+    else:
+        sentences.append(currentSentence + '.')
+        time = data['words'][x-1]['endTime']
+        endTimes.append(time)
+        currentSentence = ''
+        
+print('startTimes: ' + str(startTimes))
+print('endTimes: ' + str(startTimes))
+print(sentences)
+
+f = open("words/text.txt", "w")
+f.write("<speak><par>")
+
+for x in range(len(sentences)):
+    f.write('<media begin="{start}"><speak>{text}</speak></media>'.format(start=startTimes[x], text=sentences[x]))
+    ffmpeg += ", volume=enable='between(t,{start},{end})':volume=0.2".format(start=startTimes[x], end=endTimes[x])
+
+f.write("</par></speak>")
+f.close()
+
+final_result_url = useTemplate ('TEMPLATE_2_ID', file_path='words/text.txt', result_name='merged_video', get_url=True, fields={"target_language":target_language, "source_language":source_language, "video":video_url, "ffmpeg":ffmpeg})