Update python client after recent API chages regarding content and combined search endpoints

Author: ffont

essentia_example.py

@@ -15,7 +15,7 @@
 # Get sound info example
 print("Sound info:")
 print("-----------")
-sound = freesound_client.get_sound(96541)
+sound = freesound_client.get_sound(6)
 print("Getting sound:", sound.name)
 print("Url:", sound.url)
 print("Description:", sound.description)
@@ -26,34 +26,23 @@
 print("Sound info specifying some request parameters:")
 print("-----------")
 sound = freesound_client.get_sound(
-    96541, fields="id,name,username,duration,analysis", descriptors="lowlevel.spectral_centroid", normalized=1
-)
+    6, fields="id,name,username,duration,spectral_centroid")
 print("Getting sound:", sound.name)
 print("Username:", sound.username)
 print("Duration:", str(sound.duration), "(s)")
-print("Spectral centroid:",)
-print(sound.analysis.lowlevel.spectral_centroid.as_dict())
+print("Spectral centroid:",str(sound.spectral_centroid), "(Hz)")
 print()
 
 # Get sound analysis example
 print("Get analysis:")
 print("-------------")
 analysis = sound.get_analysis()
-
-mfcc = analysis.lowlevel.mfcc.mean
+mfcc = analysis.mfcc
 print("Mfccs:", mfcc)
 # you can also get the original json (this applies to any FreesoundObject):
 print(analysis.as_dict())
 print()
 
-# Get sound analysis example specifying some request parameters
-print("Get analysis with specific normalized descriptor:")
-print("-------------")
-analysis = sound.get_analysis(descriptors="lowlevel.spectral_centroid.mean", normalized=1)
-spectral_centroid_mean = analysis.lowlevel.spectral_centroid.mean
-print("Normalized mean of spectral centroid:", spectral_centroid_mean)
-print()
-
 # Get similar sounds example
 print("Similar sounds: ")
 print("---------------")
@@ -62,20 +51,10 @@
     print("\t-", similar_sound.name, "by", similar_sound.username)
 print()
 
-# Get similar sounds example specifying some request parameters
-print("Similar sounds specifying some request parameters:")
-print("---------------")
-results_pager = sound.get_similar(
-    page_size=10, fields="name,username", descriptors_filter="lowlevel.pitch.mean:[110 TO 180]"
-)
-for similar_sound in results_pager:
-    print("\t-", similar_sound.name, "by", similar_sound.username)
-print()
-
 # Search Example
 print("Searching for 'violoncello':")
 print("----------------------------")
-results_pager = freesound_client.text_search(
+results_pager = freesound_client.search(
     query="violoncello",
     filter="tag:tenuto duration:[1.0 TO 15.0]",
     sort="rating_desc",
@@ -91,19 +70,42 @@
     print("\t-", sound.name, "by", sound.username)
 print()
 
-# Content based search example
-print("Content based search:")
+# Search all sounds that match some audio descriptor filter
+print("Filter by audio descriptors:")
 print("---------------------")
-results_pager = freesound_client.content_based_search(
-    descriptors_filter="lowlevel.pitch.var:[* TO 20]",
-    target='lowlevel.pitch_salience.mean:1.0 lowlevel.pitch.mean:440'
+results_pager = freesound_client.search(
+    filter="spectral_centroid:[1000 TO 2000] AND pitch:[200 TO 300]",
+    fields="name,username,spectral_centroid,pitch",
 )
 
 print("Num results:", results_pager.count)
 for sound in results_pager:
-    print("\t-", sound.name, "by", sound.username)
+    print("\t-", sound.name, "by", sound.username, "with spectral centroid of", sound.spectral_centroid, "Hz and pitch of", sound.pitch, "Hz")
 print()
 
+# We can use the search endpoint for a similarity search which also filters results
+print("Searching for similar sounds and filtering by a descriptor value:")
+print("---------------")
+results_pager = freesound_client.search(
+    page_size=10, fields="name,username", similar_to=6, filter="pitch:[110 TO 180]"
+)
+for similar_sound in results_pager:
+    print("\t-", similar_sound.name, "by", similar_sound.username)
+print()
+
+# Find sounds simialar to a target vector extracted from an existing sound which might not be in Freesound
+# Note that "similarity_space" must match with the similarity space used to extract the target vector
+print("Searching for sounds similar to a target vector:")
+print("---------------")
+target_vector = [0.84835537,-0.06019006,0.35139768,-0.01221892,-0.23172645,-0.03798686,-0.03869437,0.02199453,-0.10143687,-0.03342770,0.05464298,-0.02654354,0.05424048,0.04912830,0.01449411,-0.02995046,0.04584143,0.03731462,0.06914231,-0.00702387,-0.02202889,0.01644059,-0.00153376,0.11042101,0.05432773,0.05736105,-0.03779107,-0.00909068,-0.08996461,-0.04300615,0.05610843,0.02214170,-0.02155820,0.05158299,-0.00717155,-0.04345755,-0.00519616,0.02887811,-0.02205723,0.01658933,0.02485796,-0.06228176,0.03574570,-0.04556302,-0.00497004,0.00300936,-0.01974736,-0.01391953,-0.02898939,0.01041939,-0.02836645,0.00853050,-0.03129587,0.00454572,0.00898315,-0.01371797,-0.00918297,-0.01049032,0.02800968,-0.04248178,0.02648444,-0.01034762,0.02105908,-0.01137279,0.02845560,-0.04284714,0.00797986,0.00973879,-0.00850114,-0.01093731,-0.00629640,-0.01862817,0.00829806,0.01137537,0.02601988,0.03015542,-0.01091145,0.00547907,-0.00426657,0.01001693,0.00793383,0.00082211,-0.02848534,-0.00823537,0.01392606,-0.02012341,-0.00788319,0.02797560,-0.01470957,-0.01917517,-0.01177181,0.00952904,-0.00223396,-0.01586017,-0.00566903,0.01150901,-0.00361810,-0.00257769,-0.01509761,0.00552032]
+results_pager = freesound_client.search(
+    page_size=10, fields="name,username", similar_to=target_vector, similarity_space="freesound_classic"
+)
+for similar_sound in results_pager:
+    print("\t-", similar_sound.name, "by", similar_sound.username)
+print()
+
+
 # Getting sounds from a user example
 print("User sounds:")
 print("-----------")
@@ -126,7 +128,7 @@
 user = freesound_client.get_user("Headphaze")
 print("User name:", user.username)
 results_pager = user.get_sounds(
-    page_size=10, fields="name,username,samplerate,duration,analysis", descriptors="rhythm.bpm"
+    page_size=10, fields="name,username,samplerate,duration" 
 )
 
 print("Num results:", results_pager.count)
@@ -167,7 +169,7 @@
 pack = freesound_client.get_pack(3524)
 print("Pack name:", pack.name)
 results_pager = pack.get_sounds(
-    page_size=5, fields="id,name,username,duration,analysis", descriptors="lowlevel.spectral_flatness_db"
+    page_size=5, fields="id,name,username,duration,spectral_flatness"
 )
 print("Num results:", results_pager.count)
 print("\t----- PAGE 1 -----")
@@ -181,9 +183,9 @@
     )
     print(
         sound.duration,
-        "s and a mean spectral flatness of",
+        "s and a spectral flatness of",
     )
-    print(sound.analysis.lowlevel.spectral_flatness_db.mean)
+    print(sound.spectral_flatness)
 print("\t----- PAGE 2 -----")
 results_pager = results_pager.next_page()
 for sound in results_pager:
@@ -196,7 +198,7 @@
     )
     print(
         sound.duration,
-        "s and a mean spectral flatness of",
+        "s and a spectral flatness of",
     )
-    print(sound.analysis.lowlevel.spectral_flatness_db.mean)
+    print(sound.spectral_flatness)
 print()