File size: 5,924 Bytes
ccd481a
8e0a255
1c2ef01
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bb6a23e
 
8e0a255
 
bb6a23e
8e0a255
 
 
 
bb6a23e
8e0a255
bb6a23e
8e0a255
 
 
 
bb6a23e
8e0a255
bb6a23e
 
8e0a255
 
bb6a23e
8e0a255
 
 
 
 
 
 
 
 
 
 
bb6a23e
8e0a255
 
 
 
bb6a23e
8e0a255
 
 
 
 
 
 
 
 
bb6a23e
8e0a255
 
 
 
 
 
 
bb6a23e
 
 
8e0a255
bb6a23e
8e0a255
 
bb6a23e
 
8e0a255
 
 
 
 
bb6a23e
8e0a255
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bb6a23e
8e0a255
 
 
 
 
 
 
 
bb6a23e
 
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
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
---
license: cc-by-nc-4.0
inference: False
tags:
  - audio-to-audio
  - text-to-speech
library_name: seamless_communication
language:
- af
- am
- ar
- as
- az
- be
- bn
- bs
- bg
- ca
- cs
- zh
- cy
- da
- de
- el
- en
- et
- fi
- fr
- or
- om
- ga
- gl
- gu
- ha
- he
- hi
- hr
- hu
- hy
- ig
- id
- is
- it
- jv
- ja
- kn
- ka
- kk
- mn
- km
- ky
- ko
- lo
- ln
- lt
- lb
- lg
- lv
- ml
- mr
- mk
- mt
- mi
- my
- nl
- nb
- ne
- ny
- oc
- pa
- ps
- fa
- pl
- pt
- ro
- ru
- sk
- sl
- sn
- sd
- so
- es
- sr
- sv
- sw
- ta
- te
- tg
- tl
- th
- tr
- uk
- ur
- uz
- vi
- wo
- xh
- yo
- ms
- zu
- ary
- arz
- yue
- kea
---

# SeamlessStreaming
SeamlessStreaming is a multilingual streaming translation model. It supports:

- Streaming Automatic Speech Recognition on 96 languages.
- Simultaneous translation on 101 source languages for speech input.
- Simultaneous translation on 96 target languages for text output.
- Simultaneous translation on 36 target languages for speech output.

![SeamlessStreaming architecture](streaming_arch.png)

## SeamlessStreaming  models
| Model Name         | #params | checkpoint                                                                              | metrics                                                                              |
| ------------------ | ------- | --------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------ |
| SeamlessStreaming  | 2.5B    | [🤗 Model card](https://huggingface.co/facebook/seamless-streaming) - [monotonic decoder checkpoint](https://huggingface.co/facebook/seamless-streaming/resolve/main/seamless_streaming_monotonic_decoder.pt) - [streaming UnitY2 checkpoint](https://huggingface.co/facebook/seamless-streaming/resolve/main/seamless_streaming_unity.pt)  | [metrics](https://dl.fbaipublicfiles.com/seamless/metrics/streaming/seamless_streaming.zip)  |

The evaluation data ids for FLEURS, CoVoST2 and CVSS-C can be found [here](https://dl.fbaipublicfiles.com/seamless/metrics/evaluation_data_ids.zip)


## Evaluating SeamlessStreaming models
To reproduce our results, or to evaluate using the same metrics over your own test sets, please check out the [Evaluation README here](../../src/seamless_communication/cli/streaming/README.md). Streaming evaluation depends on the SimulEval library.

## Seamless Streaming demo

### Running on HF spaces
You can simply duplicate the space to run it. [🤗 HF Space](https://huggingface.co/spaces/facebook/seamless-streaming)

## Running locally

### Install backend seamless_server dependencies

> [!NOTE]
> Please note: we *do not* recommend running the model on CPU. CPU inference will be slow and introduce noticable delays in the simultaneous translation.

> [!NOTE]
> The example below is for PyTorch stable (2.1.1) and variant cu118. 
> Check [here](https://pytorch.org/get-started/locally/) to find the torch/torchaudio command for your variant. 
> Check [here](https://github.com/facebookresearch/fairseq2#variants) to find the fairseq2 command for your variant.

If running for the first time, create conda environment and install the desired torch version. Then install the rest of the requirements:
```
cd seamless_server
conda create --yes --name smlss_server python=3.8 libsndfile==1.0.31
conda activate smlss_server
conda install --yes pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia
pip install fairseq2 --pre --extra-index-url https://fair.pkg.atmeta.com/fairseq2/whl/nightly/pt2.1.1/cu118
pip install -r requirements.txt
```

### Install frontend streaming-react-app dependencies
```
conda install -c conda-forge nodejs
cd streaming-react-app
npm install --global yarn
yarn
yarn build  # this will create the dist/ folder
```


### Running the server

The server can be run locally with uvicorn below.
Run the server in dev mode:

```
cd seamless_server
uvicorn app_pubsub:app --reload --host localhost
```

Run the server in prod mode:

```
cd seamless_server
uvicorn app_pubsub:app --host 0.0.0.0
```

To enable additional logging from uvicorn pass `--log-level debug` or `--log-level trace`.


### Debuging

If you enable "Server Debug Flag" when starting streaming from the client, this enables extensive debug logging and it saves audio files in /debug folder. 

## Citation

For EMMA, please cite :
```bibtex
@article{ma_efficient_2023,
  author={Ma, Xutai and Sun, Anna and Ouyang, Siqi and Inaguma, Hirofumi and Tomasello, Paden},
  title={Efficient Monotonic Multihead Attention},
  year={2023},
  url={https://ai.meta.com/research/publications/efficient-monotonic-multihead-attention/},
}
```

For SeamlessStreaming, please cite :
```bibtex
@inproceedings{seamless2023,
   title="Seamless: Multilingual Expressive and Streaming Speech Translation",
   author="{Seamless Communication}, Lo{\"i}c Barrault, Yu-An Chung, Mariano Coria Meglioli, David Dale, Ning Dong, Mark Duppenthaler, Paul-Ambroise Duquenne, Brian Ellis, Hady Elsahar, Justin Haaheim, John Hoffman, Min-Jae Hwang, Hirofumi Inaguma, Christopher Klaiber, Ilia Kulikov, Pengwei Li, Daniel Licht, Jean Maillard, Ruslan Mavlyutov, Alice Rakotoarison, Kaushik Ram Sadagopan, Abinesh Ramakrishnan, Tuan Tran, Guillaume Wenzek, Yilin Yang, Ethan Ye, Ivan Evtimov, Pierre Fernandez, Cynthia Gao, Prangthip Hansanti, Elahe Kalbassi, Amanda Kallet, Artyom Kozhevnikov, Gabriel Mejia, Robin San Roman, Christophe Touret, Corinne Wong, Carleigh Wood, Bokai Yu, Pierre Andrews, Can Balioglu, Peng-Jen Chen, Marta R. Costa-juss{\`a}, Maha Elbayad, Hongyu Gong, Francisco Guzm{\'a}n, Kevin Heffernan, Somya Jain, Justine Kao, Ann Lee, Xutai Ma, Alex Mourachko, Benjamin Peloquin, Juan Pino, Sravya Popuri, Christophe Ropers, Safiyyah Saleem, Holger Schwenk, Anna Sun, Paden Tomasello, Changhan Wang, Jeff Wang, Skyler Wang, Mary Williamson",
  journal={ArXiv},
  year={2023}
}
```