Hugging Face's logo

LocalAI-io
/
whisper-medium-it-multi

Automatic Speech Recognition
Safetensors
Italian
whisper
italian
localai
Model card Files
xet
 Community
whisper-medium-it-multi
File size: 2,487 Bytes
4eb3989
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
---
language: it
license: mit
tags:
- whisper
- automatic-speech-recognition
- italian
- localai
datasets:
- mozilla-foundation/common_voice_25_0
- facebook/multilingual_librispeech
- facebook/voxpopuli
base_model: openai/whisper-medium
pipeline_tag: automatic-speech-recognition
---

# whisper-medium-it-multi

Fine-tuned [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) (769M params) for Italian ASR on multiple datasets.

**Author:** Ettore Di Giacinto

Brought to you by the [LocalAI](https://github.com/mudler/LocalAI) team. This model can be used directly with [LocalAI](https://localai.io).

## Usage with LocalAI

This model is ready to use with [LocalAI](https://localai.io) via the `whisperx` backend.

Save the following as `whisperx-medium-it-multi.yaml` in your LocalAI models directory:

```yaml
name: whisperx-medium-it-multi
backend: whisperx
known_usecases:
  - transcript
parameters:
  model: LocalAI-io/whisper-medium-it-multi-ct2-int8
  language: it
```

Then transcribe audio via the OpenAI-compatible endpoint:

```bash
curl http://localhost:8080/v1/audio/transcriptions \
  -H "Content-Type: multipart/form-data" \
  -F file="@audio.mp3" \
  -F model="whisperx-medium-it-multi"
```

## Results

Evaluated on combined test set (Common Voice + MLS + VoxPopuli):

| Step | WER |
|------|-----|
| 1000 | 17.55% |
| 3000 | 16.71% |
| 5000 | 14.00% |
| 7000 | 13.02% |
| 9000 | 12.10% |
| 10000 | **12.37%** |

## Training Details

- **Base model:** openai/whisper-medium (769M parameters)
- **Datasets:** Common Voice 25.0 Italian (173k) + MLS Italian (60k) + VoxPopuli Italian (23k) = 255k train samples
- **Steps:** 10,000
- **Precision:** bf16 on NVIDIA GB10

## Usage

### Transformers

```python
from transformers import pipeline

pipe = pipeline("automatic-speech-recognition", model="LocalAI-io/whisper-medium-it-multi")
result = pipe("audio.mp3", generate_kwargs={"language": "it", "task": "transcribe"})
print(result["text"])
```

### CTranslate2 / faster-whisper

For optimized CPU inference: [LocalAI-io/whisper-medium-it-multi-ct2-int8](https://huggingface.co/LocalAI-io/whisper-medium-it-multi-ct2-int8)

## Links

- **CTranslate2 INT8:** [LocalAI-io/whisper-medium-it-multi-ct2-int8](https://huggingface.co/LocalAI-io/whisper-medium-it-multi-ct2-int8)
- **Project:** [github.com/localai-org/italian-whisper](https://github.com/localai-org/italian-whisper)
- **LocalAI:** [github.com/mudler/LocalAI](https://github.com/mudler/LocalAI)