nvidia

/

parakeet-tdt-0.6b-v3

like 795

Follow

NVIDIA 55.2k

Please developer fix this error.
=== 3. Memulai Transkripsi (145 bagian) ===
Proses Transkripsi: 0%| | 0/145 [00:00<?, ?bagian/s][NeMo I 2026-02-18 08:11:59 nemo_logging:393] Timestamps requested, setting decoding timestamps to True. Capture them in Hypothesis object, with output[0][idx].timestep['word'/'segment'/'char']
[NeMo I 2026-02-18 08:11:59 nemo_logging:393] Using RNNT Loss : tdt
Loss tdt_kwargs: {'fastemit_lambda': 0.0, 'clamp': -1.0, 'durations': [0, 1, 2, 3, 4], 'sigma': 0.02, 'omega': 0.1}
[NeMo W 2026-02-18 08:11:59 nemo_logging:405] The following configuration keys are ignored by Lhotse dataloader: use_start_end_token
[NeMo W 2026-02-18 08:11:59 nemo_logging:405] You are using a non-tarred dataset and requested tokenization during data sampling (pretokenize=True). This will cause the tokenization to happen in the main (GPU) process,possibly impacting the training speed if your tokenizer is very large.If the impact is noticable, set pretokenize=False in dataloader config.(note: that will disable token-per-second filtering and 2D bucketing features)
Proses Transkripsi: 0%| | 0/145 [00:00<?, ?bagian/s] ⚠️ Error pada bagian 1: not enough values to unpack (expected 6, got 5)

UnboundLocalError Traceback (most recent call last)
/tmp/ipython-input-1256625449.py in <cell line: 0>()
130
131 if os.path.exists(file_audio_anda):
--> 132 speech_to_text_with_progress(file_audio_anda)
133 else:
134 print(f"❌ File '{file_audio_anda}' tidak ditemukan. Mohon upload file terlebih dahulu.")

/tmp/ipython-input-1256625449.py in speech_to_text_with_progress(input_file, chunk_length_ms)
94 if os.path.exists(chunk_filename):
95 os.remove(chunk_filename)
---> 96 del output
97 torch.cuda.empty_cache()
98 gc.collect()

UnboundLocalError: cannot access local variable 'output' where it is not associated with a value