Automatic Speech Recognition
Transformers
Safetensors
phi4mm
text-generation
nlp
code
audio
speech-summarization
speech-translation
visual-question-answering
phi-4-multimodal
phi
phi-4-mini
custom_code
Eval Results
Instructions to use microsoft/Phi-4-multimodal-instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use microsoft/Phi-4-multimodal-instruct with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="microsoft/Phi-4-multimodal-instruct", trust_remote_code=True)# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("microsoft/Phi-4-multimodal-instruct", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Add Open ASR Leaderboard evaluation results
#93
by SaylorTwift HF Staff - opened
.eval_results/open_asr_leaderboard.yaml
ADDED
|
@@ -0,0 +1,99 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
- dataset:
|
| 2 |
+
id: hf-audio/open-asr-leaderboard
|
| 3 |
+
task_id: mean_wer
|
| 4 |
+
value: 6.02
|
| 5 |
+
date: '2025-02-24'
|
| 6 |
+
source:
|
| 7 |
+
url: https://huggingface.co/hf-audio
|
| 8 |
+
name: open-asr-leaderboard
|
| 9 |
+
user: hf-audio
|
| 10 |
+
|
| 11 |
+
- dataset:
|
| 12 |
+
id: hf-audio/open-asr-leaderboard
|
| 13 |
+
task_id: rtfx
|
| 14 |
+
value: 151.1
|
| 15 |
+
date: '2025-02-24'
|
| 16 |
+
source:
|
| 17 |
+
url: https://huggingface.co/hf-audio
|
| 18 |
+
name: open-asr-leaderboard
|
| 19 |
+
user: hf-audio
|
| 20 |
+
|
| 21 |
+
- dataset:
|
| 22 |
+
id: hf-audio/open-asr-leaderboard
|
| 23 |
+
task_id: ami_wer
|
| 24 |
+
value: 11.09
|
| 25 |
+
date: '2025-02-24'
|
| 26 |
+
source:
|
| 27 |
+
url: https://huggingface.co/hf-audio
|
| 28 |
+
name: open-asr-leaderboard
|
| 29 |
+
user: hf-audio
|
| 30 |
+
|
| 31 |
+
- dataset:
|
| 32 |
+
id: hf-audio/open-asr-leaderboard
|
| 33 |
+
task_id: earnings22_wer
|
| 34 |
+
value: 10.16
|
| 35 |
+
date: '2025-02-24'
|
| 36 |
+
source:
|
| 37 |
+
url: https://huggingface.co/hf-audio
|
| 38 |
+
name: open-asr-leaderboard
|
| 39 |
+
user: hf-audio
|
| 40 |
+
|
| 41 |
+
- dataset:
|
| 42 |
+
id: hf-audio/open-asr-leaderboard
|
| 43 |
+
task_id: gigaspeech_wer
|
| 44 |
+
value: 9.33
|
| 45 |
+
date: '2025-02-24'
|
| 46 |
+
source:
|
| 47 |
+
url: https://huggingface.co/hf-audio
|
| 48 |
+
name: open-asr-leaderboard
|
| 49 |
+
user: hf-audio
|
| 50 |
+
|
| 51 |
+
- dataset:
|
| 52 |
+
id: hf-audio/open-asr-leaderboard
|
| 53 |
+
task_id: librispeech_clean_wer
|
| 54 |
+
value: 1.69
|
| 55 |
+
date: '2025-02-24'
|
| 56 |
+
source:
|
| 57 |
+
url: https://huggingface.co/hf-audio
|
| 58 |
+
name: open-asr-leaderboard
|
| 59 |
+
user: hf-audio
|
| 60 |
+
|
| 61 |
+
- dataset:
|
| 62 |
+
id: hf-audio/open-asr-leaderboard
|
| 63 |
+
task_id: librispeech_other_wer
|
| 64 |
+
value: 3.82
|
| 65 |
+
date: '2025-02-24'
|
| 66 |
+
source:
|
| 67 |
+
url: https://huggingface.co/hf-audio
|
| 68 |
+
name: open-asr-leaderboard
|
| 69 |
+
user: hf-audio
|
| 70 |
+
|
| 71 |
+
- dataset:
|
| 72 |
+
id: hf-audio/open-asr-leaderboard
|
| 73 |
+
task_id: spgispeech_wer
|
| 74 |
+
value: 3.06
|
| 75 |
+
date: '2025-02-24'
|
| 76 |
+
source:
|
| 77 |
+
url: https://huggingface.co/hf-audio
|
| 78 |
+
name: open-asr-leaderboard
|
| 79 |
+
user: hf-audio
|
| 80 |
+
|
| 81 |
+
- dataset:
|
| 82 |
+
id: hf-audio/open-asr-leaderboard
|
| 83 |
+
task_id: tedlium_wer
|
| 84 |
+
value: 2.94
|
| 85 |
+
date: '2025-02-24'
|
| 86 |
+
source:
|
| 87 |
+
url: https://huggingface.co/hf-audio
|
| 88 |
+
name: open-asr-leaderboard
|
| 89 |
+
user: hf-audio
|
| 90 |
+
|
| 91 |
+
- dataset:
|
| 92 |
+
id: hf-audio/open-asr-leaderboard
|
| 93 |
+
task_id: voxpopuli_wer
|
| 94 |
+
value: 6.04
|
| 95 |
+
date: '2025-02-24'
|
| 96 |
+
source:
|
| 97 |
+
url: https://huggingface.co/hf-audio
|
| 98 |
+
name: open-asr-leaderboard
|
| 99 |
+
user: hf-audio
|