file ids - transcriptions

spgispeech.py

@@ -77,9 +77,9 @@ _META_FILENAMES = {
     "test": "test.csv",
 }
 
-_WHISPER_TRANSCRIPT_URL = "https://huggingface.co/datasets/distil-whisper/spgispeech/resolve/main/transcription_data/greedy_search/"
+_WHISPER_TRANSCRIPT_URL = "https://huggingface.co/datasets/distil-whisper/whisper_transcriptions_greedy/resolve/main/spgispeech"
 
-_WHISPER_TRANSCRIPT_URLs = _WHISPER_TRANSCRIPT_URL + "/{split}-transcription.txt"
+_WHISPER_TRANSCRIPT_URLs = _WHISPER_TRANSCRIPT_URL + "/{split}-transcription.csv"
 
 
 class Spgispeech(datasets.GeneratorBasedBuilder):
@@ -208,13 +208,11 @@ class Spgispeech(datasets.GeneratorBasedBuilder):
             csvreader = csv.DictReader(f, delimiter="|")
             metadata = {x["wav_filename"]: dict((k, x[k]) for k in dict_keys) for x in csvreader}
 
-        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 reader:
+                whisper_transcriptions[line["file_id"]] = line["whisper_transcript"]
 
         for local_extracted_archive_path, archive in zip(local_extracted_archive_paths, archives):
             # Here we iterate over all the files within the TAR archive:
@@ -233,6 +231,5 @@ class Spgispeech(datasets.GeneratorBasedBuilder):
                 example["id"] = example.pop("wav_filename")
                 example["text"] = example.pop("transcript")
                 example["audio"] = {"path": path, "bytes": audio_file.read()}
-                example["whisper_transcript"] = whisper_transcripts[idx]
+                example["whisper_transcript"] = whisper_transcriptions.get(example["id"], None)
                 yield audio_filename, example
-                idx += 1