file ids - transcriptions

common_voice_13_0.py

@@ -50,9 +50,9 @@ _TRANSCRIPT_URL = _BASE_URL + "transcript/{lang}/{split}.tsv"
 
 _N_SHARDS_URL = _BASE_URL + "n_shards.json"
 
-_WHISPER_TRANSCRIPT_URL = "https://huggingface.co/datasets/distil-whisper/common_voice_13_0/resolve/main/transcription_data/greedy_search/"
+_WHISPER_TRANSCRIPT_URL = "https://huggingface.co/datasets/distil-whisper/whisper_transcriptions_greedy/resolve/main/common_voice_13_0"
 
-_WHISPER_TRANSCRIPT_URLs = _WHISPER_TRANSCRIPT_URL + "/{split}-transcription.txt"
+_WHISPER_TRANSCRIPT_URLs = _WHISPER_TRANSCRIPT_URL + "/{split}-transcription.csv"
 
 
 class CommonVoiceConfig(datasets.BuilderConfig):
@@ -110,6 +110,7 @@ class CommonVoice(datasets.GeneratorBasedBuilder):
         features = datasets.Features(
             {
                 "client_id": datasets.Value("string"),
+                "path": datasets.Value("string"),
                 "id": datasets.Value("string"),
                 "audio": datasets.features.Audio(sampling_rate=48_000),
                 "text": datasets.Value("string"),
@@ -196,13 +197,11 @@ class CommonVoice(datasets.GeneratorBasedBuilder):
                         row[field] = ""
                 metadata[row["path"]] = row
 
-        whisper_transcripts = []
-
+        whisper_transcriptions = dict()
         with open(whisper_transcript, encoding="utf-8") as f:
-            for row in f:
-                whisper_transcripts.append(row.rstrip("\n"))
-
-        idx = 0
+            reader = csv.DictReader(f, delimiter=",", quoting=csv.QUOTE_NONE)
+            for line in tqdm(reader, desc="Reading transcriptions..."):
+                whisper_transcriptions[line["file_id"]] = line["whisper_transcript"]
 
         for i, audio_archive in enumerate(archives):
             for filename, file in audio_archive:
@@ -213,9 +212,8 @@ class CommonVoice(datasets.GeneratorBasedBuilder):
                     # set the audio feature and the path to the extracted file
                     path = os.path.join(local_extracted_archive_paths[i], filename) if local_extracted_archive_paths else filename
                     result["audio"] = {"path": path, "bytes": file.read()}
+                    result["path"] = path
                     result["id"] = filename
-                    result["whisper_transcript"] = whisper_transcripts[idx]
+                    result["whisper_transcript"] = whisper_transcriptions.get(filename, None)
 
                     yield path, result
-
-                    idx += 1