|
|
@@ -12,13 +12,13 @@ SeamlessM4T models support the tasks of:
|
|
|
|
|
|
:star2: We are releasing SemalessM4T v2, an updated version with our novel *UnitY2* architecture. This new model improves over SeamlessM4T v1 in quality as well as inference latency in speech generation tasks.
|
|
|
|
|
|
-To learn more about the collection of SeamlessM4T models, the approach used in each, their language coverage and their performance, visit the [SeamlessM4T README](docs/m4t/README.md)
|
|
|
+To learn more about the collection of SeamlessM4T models, the approach used in each, their language coverage and their performance, visit the [SeamlessM4T README](docs/m4t/README.md) or [🤗 Model Card](https://huggingface.co/facebook/seamless-m4t-v2-large)
|
|
|
|
|
|
## SeamlessExpressive
|
|
|
|
|
|
SeamlessExpressive is a speech-to-speech translation model that captures certain underexplored aspects of prosody such as speech rate and pauses, while preserving the style of one's voice and high content translation quality.
|
|
|
|
|
|
-To learn more about SeamlessExpressive models, visit the [SeamlessExpressive README](docs/expressive/README.md)
|
|
|
+To learn more about SeamlessExpressive models, visit the [SeamlessExpressive README](docs/expressive/README.md) or [🤗 Model Card](https://huggingface.co/facebook/seamless-expressive)
|
|
|
|
|
|
|
|
|
## SeamlessStreaming
|
|
|
@@ -30,7 +30,7 @@ The SeamlessStreaming model supports the following tasks:
|
|
|
- Speech-to-text translation (S2TT)
|
|
|
- Automatic speech recognition (ASR)
|
|
|
|
|
|
-To learn more about SeamlessStreaming models, visit the [SeamlessStreaming README](docs/streaming/README.md)
|
|
|
+To learn more about SeamlessStreaming models, visit the [SeamlessStreaming README](docs/streaming/README.md) or [🤗 Model Card](https://huggingface.co/facebook/seamless-streaming)
|
|
|
|
|
|
## Seamless
|
|
|
|
|
|
@@ -137,7 +137,7 @@ python app.py
|
|
|
### SeamlessM4T models
|
|
|
| Model Name | #params | checkpoint | metrics |
|
|
|
| ----------------------- | ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ----------------------------------------------------------------------------------- |
|
|
|
-| SeamlessM4T-Large v2 | 2.3B | [🤗 Model card](https://huggingface.co/facebook/seamless-m4t-v2-large) - [checkpoint](https://dl.fbaipublicfiles.com/seamless/models/seamlessM4T_v2_large.pt) | [metrics](https://dl.fbaipublicfiles.com/seamless/metrics/seamlessM4T_large_v2.zip) |
|
|
|
+| SeamlessM4T-Large v2 | 2.3B | [🤗 Model card](https://huggingface.co/facebook/seamless-m4t-v2-large) - [checkpoint](https://huggingface.co/facebook/seamless-m4t-v2-large/resolve/main/seamlessM4T_v2_large.pt ) | [metrics](https://dl.fbaipublicfiles.com/seamless/metrics/seamlessM4T_large_v2.zip) |
|
|
|
| SeamlessM4T-Large (v1) | 2.3B | [🤗 Model card](https://huggingface.co/facebook/seamless-m4t-large) - [checkpoint](https://huggingface.co/facebook/seamless-m4t-large/resolve/main/multitask_unity_large.pt) | [metrics](https://dl.fbaipublicfiles.com/seamless/metrics/seamlessM4T_large.zip) |
|
|
|
| SeamlessM4T-Medium (v1) | 1.2B | [🤗 Model card](https://huggingface.co/facebook/seamless-m4t-medium) - [checkpoint](https://huggingface.co/facebook/seamless-m4t-medium/resolve/main/multitask_unity_medium.pt) | [metrics](https://dl.fbaipublicfiles.com/seamless/metrics/seamlessM4T_medium.zip) |
|
|
|
|
|
|
@@ -152,7 +152,7 @@ Please note that SeamlessExpressive is made available under its own [License](SE
|
|
|
### SeamlessStreaming models
|
|
|
| Model Name | #params | checkpoint | metrics |
|
|
|
| ----------------- | ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------- |
|
|
|
-| SeamlessStreaming | 2.5B | [🤗 Model card](https://huggingface.co/facebook/seamless-streaming) - [monotonic decoder checkpoint](https://dl.fbaipublicfiles.com/seamless/models/seamless_streaming_monotonic_decoder.pt) - [streaming UnitY2 checkpoint](https://dl.fbaipublicfiles.com/seamless/models/seamless_streaming_unity.pt) | [metrics](https://dl.fbaipublicfiles.com/seamless/metrics/streaming/seamless_streaming.zip) |
|
|
|
+| 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) |
|
|
|
|
|
|
### Seamless models
|
|
|
Seamless model is simply the SeamlessStreaming model with the non-expressive `vocoder_v2` swapped out with the expressive `vocoder_pretssel`.
|
Vaibhav Srivastav