Downloads · 30 days
27
4% of all-time downloads
Peacockery/parakeet-ctc-109m-farsi
parakeet-ctc-109m-farsi is a automatic speech recognition model from Peacockery. Use it when you need speech turned into text. It is set up for nemo. The card lists the license as cc-by-4.0.
A Farsi fine-tune of NVIDIA's Parakeet CTC 109M (the CTC branch of parakeet-tdtctc-110m), trained with a custom Farsi SentencePiece tokenizer on 844 hours of quality-classified Farsi speech. It is the best Parakeet-fa…
Downloads · 30 days
27
4% of all-time downloads
All-time downloads
646
Public
Repo size
438 MB
Likes
0
Public
Click a slice to open those files.
.nemo438 MB · 100%
From the Hugging Face model README
A Farsi fine-tune of NVIDIA's Parakeet CTC 109M (the CTC branch of parakeet-tdt_ctc-110m), trained with a custom Farsi SentencePiece tokenizer on 844 hours of quality-classified Farsi speech. It is the best Parakeet-family model from this project; a later variant trained on a stricter 224-hour subset regressed and was retired.
model.nemo: the complete NeMo checkpoint (weights + tokenizer + config)844 hours of Farsi speech: Common Voice, Thomcles, Mana TTS, Neyshekar, FLEURS, WorldSpeech, and Farsi YouTube, with web-scraped portions machine-labeled and filtered by a Scribe quality classifier. Clips capped at 400 characters and 60 chars/sec to drop misaligned transcripts.
Test splits, corpus-level jiwer after shared normalization, re-scored 2026-05-28 (WER %):
| split | WER | CER |
|---|---|---|
| FLEURS | 14.94 | 4.15 |
| Common Voice 25 | 23.19 | 6.27 |
| Mana TTS | 28.53 | 5.07 |
| Neyshekar | 27.87 | 5.48 |
| WorldSpeech | 40.12 | 20.51 |
| YouTube | 34.55 | 13.72 |
For a stronger Farsi model from the same project see Peacockery/omni-ctc-300m-farsi, which wins on every split. This checkpoint is the lighter and faster option (109M parameters, ~350x realtime on the benchmark hardware).
import nemo.collections.asr as nemo_asr
model = nemo_asr.models.EncDecCTCModelBPE.restore_from("model.nemo")
texts = model.transcribe(["audio.wav"])