Automatic Speech Recognition
NeMo
English
speech
multi-talker-asr
target-speaker-asr
speaker-diarization
meeting-transcription
long-form
parakeet
fastconformer
ctc
Instructions to use BUT-FIT/Pandemonium with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- NeMo
How to use BUT-FIT/Pandemonium with NeMo:
import nemo.collections.asr as nemo_asr asr_model = nemo_asr.models.ASRModel.from_pretrained("BUT-FIT/Pandemonium") transcriptions = asr_model.transcribe(["file.wav"]) - Notebooks
- Google Colab
- Kaggle
Correct the description of the pseudo-labeled stage
Browse files
README.md
CHANGED
|
@@ -11,6 +11,10 @@ pipeline_tag: automatic-speech-recognition
|
|
| 11 |
library_name: nemo
|
| 12 |
base_model: nvidia/parakeet-tdt-0.6b-v3
|
| 13 |
tags:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
- speech
|
| 15 |
- automatic-speech-recognition
|
| 16 |
- multi-talker-asr
|
|
@@ -83,7 +87,7 @@ Three stages, each starting from the previous one:
|
|
| 83 |
| stage | data | hours |
|
| 84 |
|---|---|---|
|
| 85 |
| 1. synthetic | [FastMSS](https://arxiv.org/abs/2605.15442) mixtures of up to 8 LibriSpeech speakers (2k h) + Fisher conversations restored with Sidon and force-aligned (2k h) | 4.0k |
|
| 86 |
-
| 2. pseudo-labeled | multi-speaker YODAS2 conversations
|
| 87 |
| 3. real | NOTSOFAR-1 and AMI single-distant-microphone training sets | 115 |
|
| 88 |
|
| 89 |
Training conversations are capped at 8 speakers. The diarization mask is randomly corrupted during
|
|
|
|
| 11 |
library_name: nemo
|
| 12 |
base_model: nvidia/parakeet-tdt-0.6b-v3
|
| 13 |
tags:
|
| 14 |
+
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+
|
| 18 |
- speech
|
| 19 |
- automatic-speech-recognition
|
| 20 |
- multi-talker-asr
|
|
|
|
| 87 |
| stage | data | hours |
|
| 88 |
|---|---|---|
|
| 89 |
| 1. synthetic | [FastMSS](https://arxiv.org/abs/2605.15442) mixtures of up to 8 LibriSpeech speakers (2k h) + Fisher conversations restored with Sidon and force-aligned (2k h) | 4.0k |
|
| 90 |
+
| 2. pseudo-labeled | multi-speaker YODAS2 conversations in the 23 languages Parakeet-v3 supports, automatically diarized and transcribed by DiCoW v3.3 conditioned on that diarization | 4.6k |
|
| 91 |
| 3. real | NOTSOFAR-1 and AMI single-distant-microphone training sets | 115 |
|
| 92 |
|
| 93 |
Training conversations are capped at 8 speakers. The diarization mask is randomly corrupted during
|