url
stringlengths 23
7.17k
| text
stringlengths 0
1.65M
|
|---|---|
https://huggingface.co/spaces/awacke1/Memory-Shared | This Space is sleeping due to inactivity. |
https://huggingface.co/averma26 | Abhishek Verma
averma26
Research interests
NLP Learning about Large Language Model Would like to learn about RLHF Multi-modal Applications
Organizations |
https://huggingface.co/spaces/awacke1/AI.Dashboard.HEDIS | App Files Files Community |
https://huggingface.co/spaces/awacke1/Model-Easy-Button-Generative-Images-runwayml-stable-diffusion-v1-5 | runtime error
failed to create containerd task: failed to create shim task: OCI runtime create failed: runc create failed: unable to create new parent process: namespace path: lstat /proc/0/ns/ipc: no such file or directory: unknown
Container logs: |
https://huggingface.co/spaces/awacke1/SaveAndReloadDataset | runtime error
Traceback (most recent call last): File "/home/user/.local/bin/streamlit", line 5, in <module> from streamlit.cli import main File "/home/user/.local/lib/python3.8/site-packages/streamlit/__init__.py", line 70, in <module> from streamlit.delta_generator import DeltaGenerator as _DeltaGenerator File "/home/user/.local/lib/python3.8/site-packages/streamlit/delta_generator.py", line 73, in <module> from streamlit.elements.arrow_altair import ArrowAltairMixin File "/home/user/.local/lib/python3.8/site-packages/streamlit/elements/arrow_altair.py", line 25, in <module> from altair.vegalite.v4.api import Chart ModuleNotFoundError: No module named 'altair.vegalite.v4'
Container logs: |
https://huggingface.co/spaces/awacke1/Health-Care-AI-and-Datasets | runtime error
2023-09-26 00:32:39.483 INFO matplotlib.font_manager: generated new fontManager Traceback (most recent call last): File "/home/user/.local/bin/streamlit", line 5, in <module> from streamlit.web.cli import main File "/home/user/.local/lib/python3.8/site-packages/streamlit/__init__.py", line 55, in <module> from streamlit.delta_generator import DeltaGenerator as _DeltaGenerator File "/home/user/.local/lib/python3.8/site-packages/streamlit/delta_generator.py", line 45, in <module> from streamlit.elements.arrow_altair import ArrowAltairMixin File "/home/user/.local/lib/python3.8/site-packages/streamlit/elements/arrow_altair.py", line 36, in <module> from altair.vegalite.v4.api import Chart ModuleNotFoundError: No module named 'altair.vegalite.v4'
Container logs: |
https://huggingface.co/Classroom-workshop/assignment2-francesco | Edit model card
PPO Agent playing LunarLander-v2
This is a trained model of a PPO agent playing LunarLander-v2 using the stable-baselines3 library.
Usage (with Stable-baselines3)
TODO: Add your code
Evaluation results
mean_reward on LunarLander-v2
self-reported
311.40 +/- 10.16 |
https://huggingface.co/spaces/awacke1/chatgpt-demo | This Space is sleeping due to inactivity. |
https://huggingface.co/spaces/awacke1/CSVDatasetAnalyzer | runtime error
Scheduling failure: not enough hardware capacity
Container logs:
Fetching error logs... |
https://huggingface.co/Classroom-workshop/assignment2-llama | Edit model card
PPO Agent playing LunarLander-v2
This is a trained model of a PPO agent playing LunarLander-v2 using the stable-baselines3 library.
Usage (with Stable-baselines3)
TODO: Add your code
Evaluation results
mean_reward on LunarLander-v2
self-reported
200.68 +/- 7.11 |
https://huggingface.co/login?next=%2FClassroom-workshop | Log In
Don't have an account? Sign Up
Username or Email address Password
Forgot your password?
SSO is available for companies |
https://huggingface.co/Classroom-workshop/assignment2-thom | Edit model card
PPO Agent playing LunarLander-v2
This is a trained model of a PPO agent playing LunarLander-v2 using the stable-baselines3 library.
Usage (with Stable-baselines3)
TODO: Add your code
Evaluation results
mean_reward on LunarLander-v2
self-reported
0.000 |
https://huggingface.co/Classroom-workshop/assignment1-joane | S2T-SMALL-LIBRISPEECH-ASR
s2t-small-librispeech-asr is a Speech to Text Transformer (S2T) model trained for automatic speech recognition (ASR). The S2T model was proposed in this paper and released in this repository
Model description
S2T is an end-to-end sequence-to-sequence transformer model. It is trained with standard autoregressive cross-entropy loss and generates the transcripts autoregressively.
Intended uses & limitations
This model can be used for end-to-end speech recognition (ASR). See the model hub to look for other S2T checkpoints.
How to use
As this a standard sequence to sequence transformer model, you can use the generate method to generate the transcripts by passing the speech features to the model.
Note: The Speech2TextProcessor object uses torchaudio to extract the filter bank features. Make sure to install the torchaudio package before running this example.
Note: The feature extractor depends on torchaudio and the tokenizer depends on sentencepiece so be sure to install those packages before running the examples.
You could either install those as extra speech dependancies with pip install transformers"[speech, sentencepiece]" or install the packages seperatly with pip install torchaudio sentencepiece.
import torch from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration from datasets import load_dataset model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr") processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr") ds = load_dataset( "patrickvonplaten/librispeech_asr_dummy", "clean", split="validation" ) input_features = processor( ds[0]["audio"]["array"], sampling_rate=16_000, return_tensors="pt" ).input_features # Batch size 1 generated_ids = model.generate(input_ids=input_features) transcription = processor.batch_decode(generated_ids)
Evaluation on LibriSpeech Test
The following script shows how to evaluate this model on the LibriSpeech "clean" and "other" test dataset.
from datasets import load_dataset, load_metric from transformers import Speech2TextForConditionalGeneration, Speech2TextProcessor librispeech_eval = load_dataset("librispeech_asr", "clean", split="test") # change to "other" for other test dataset wer = load_metric("wer") model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr").to("cuda") processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr", do_upper_case=True) librispeech_eval = librispeech_eval.map(map_to_array) def map_to_pred(batch): features = processor(batch["audio"]["array"], sampling_rate=16000, padding=True, return_tensors="pt") input_features = features.input_features.to("cuda") attention_mask = features.attention_mask.to("cuda") gen_tokens = model.generate(input_ids=input_features, attention_mask=attention_mask) batch["transcription"] = processor.batch_decode(gen_tokens, skip_special_tokens=True) return batch result = librispeech_eval.map(map_to_pred, batched=True, batch_size=8, remove_columns=["speech"]) print("WER:", wer(predictions=result["transcription"], references=result["text"]))
Result (WER):
Training data
The S2T-SMALL-LIBRISPEECH-ASR is trained on LibriSpeech ASR Corpus, a dataset consisting of approximately 1000 hours of 16kHz read English speech.
Training procedure
Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization) is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 10,000.
Training
The model is trained with standard autoregressive cross-entropy loss and using SpecAugment. The encoder receives speech features, and the decoder generates the transcripts autoregressively.
BibTeX entry and citation info
@inproceedings{wang2020fairseqs2t, title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq}, author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino}, booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations}, year = {2020}, } |
https://huggingface.co/Classroom-workshop/assignment1-maria | S2T-SMALL-LIBRISPEECH-ASR
s2t-small-librispeech-asr is a Speech to Text Transformer (S2T) model trained for automatic speech recognition (ASR). The S2T model was proposed in this paper and released in this repository
Model description
S2T is an end-to-end sequence-to-sequence transformer model. It is trained with standard autoregressive cross-entropy loss and generates the transcripts autoregressively.
Intended uses & limitations
This model can be used for end-to-end speech recognition (ASR). See the model hub to look for other S2T checkpoints.
How to use
As this a standard sequence to sequence transformer model, you can use the generate method to generate the transcripts by passing the speech features to the model.
Note: The Speech2TextProcessor object uses torchaudio to extract the filter bank features. Make sure to install the torchaudio package before running this example.
Note: The feature extractor depends on torchaudio and the tokenizer depends on sentencepiece so be sure to install those packages before running the examples.
You could either install those as extra speech dependancies with pip install transformers"[speech, sentencepiece]" or install the packages seperatly with pip install torchaudio sentencepiece.
import torch from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration from datasets import load_dataset model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr") processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr") ds = load_dataset( "patrickvonplaten/librispeech_asr_dummy", "clean", split="validation" ) input_features = processor( ds[0]["audio"]["array"], sampling_rate=16_000, return_tensors="pt" ).input_features # Batch size 1 generated_ids = model.generate(input_ids=input_features) transcription = processor.batch_decode(generated_ids)
Evaluation on LibriSpeech Test
The following script shows how to evaluate this model on the LibriSpeech "clean" and "other" test dataset.
from datasets import load_dataset, load_metric from transformers import Speech2TextForConditionalGeneration, Speech2TextProcessor librispeech_eval = load_dataset("librispeech_asr", "clean", split="test") # change to "other" for other test dataset wer = load_metric("wer") model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr").to("cuda") processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr", do_upper_case=True) librispeech_eval = librispeech_eval.map(map_to_array) def map_to_pred(batch): features = processor(batch["audio"]["array"], sampling_rate=16000, padding=True, return_tensors="pt") input_features = features.input_features.to("cuda") attention_mask = features.attention_mask.to("cuda") gen_tokens = model.generate(input_ids=input_features, attention_mask=attention_mask) batch["transcription"] = processor.batch_decode(gen_tokens, skip_special_tokens=True) return batch result = librispeech_eval.map(map_to_pred, batched=True, batch_size=8, remove_columns=["speech"]) print("WER:", wer(predictions=result["transcription"], references=result["text"]))
Result (WER):
Training data
The S2T-SMALL-LIBRISPEECH-ASR is trained on LibriSpeech ASR Corpus, a dataset consisting of approximately 1000 hours of 16kHz read English speech.
Training procedure
Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization) is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 10,000.
Training
The model is trained with standard autoregressive cross-entropy loss and using SpecAugment. The encoder receives speech features, and the decoder generates the transcripts autoregressively.
BibTeX entry and citation info
@inproceedings{wang2020fairseqs2t, title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq}, author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino}, booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations}, year = {2020}, } |
https://huggingface.co/spaces/awacke1/Prompt-Refinery-Text-to-Image-Generation/discussions | Welcome to the community
The community tab is the place to discuss and collaborate with the HF community! |
https://huggingface.co/Classroom-workshop/assignment2-omar | Edit model card
PPO Agent playing LunarLander-v2
This is a trained model of a PPO agent playing LunarLander-v2 using the stable-baselines3 library.
Usage (with Stable-baselines3)
TODO: Add your code
Evaluation results
mean_reward on LunarLander-v2
self-reported
10 +/- 7.11 |
https://huggingface.co/login?next=%2FIcelandAI | Log In
Don't have an account? Sign Up
Username or Email address Password
Forgot your password?
SSO is available for companies |
https://huggingface.co/spaces/awacke1/acw-dr-llama-7b-chat/discussions | Welcome to the community
The community tab is the place to discuss and collaborate with the HF community! |
https://huggingface.co/spaces/awacke1/Prompt-Refinery-Text-to-Image-Generation/tree/main | awacke1
Update README.md
7544130 about 2 months ago |
https://huggingface.co/spaces/IcelandAI/AnimalsOfIceland | runtime error
Traceback (most recent call last): File "/home/user/.local/bin/streamlit", line 5, in <module> from streamlit.web.cli import main File "/home/user/.local/lib/python3.8/site-packages/streamlit/__init__.py", line 55, in <module> from streamlit.delta_generator import DeltaGenerator as _DeltaGenerator File "/home/user/.local/lib/python3.8/site-packages/streamlit/delta_generator.py", line 45, in <module> from streamlit.elements.arrow_altair import ArrowAltairMixin File "/home/user/.local/lib/python3.8/site-packages/streamlit/elements/arrow_altair.py", line 36, in <module> from altair.vegalite.v4.api import Chart ModuleNotFoundError: No module named 'altair.vegalite.v4'
Container logs: |
https://huggingface.co/login?next=%2FAIPairProgramming | Log In
Don't have an account? Sign Up
Username or Email address Password
Forgot your password?
SSO is available for companies |
https://huggingface.co/Classroom-workshop/assignment2-julien | Edit model card
PPO Agent playing LunarLander-v2
This is a trained model of a PPO agent playing LunarLander-v2 using the stable-baselines3 library.
Usage (with Stable-baselines3)
TODO: Add your code
Evaluation results
mean_reward on LunarLander-v2
self-reported
0 +/- 10.16 |
https://huggingface.co/spaces/IcelandAI/Iceland-Top-Ten-Things-To-See | runtime error
Traceback (most recent call last): File "/home/user/.local/bin/streamlit", line 5, in <module> from streamlit.web.cli import main File "/home/user/.local/lib/python3.8/site-packages/streamlit/__init__.py", line 55, in <module> from streamlit.delta_generator import DeltaGenerator as _DeltaGenerator File "/home/user/.local/lib/python3.8/site-packages/streamlit/delta_generator.py", line 45, in <module> from streamlit.elements.arrow_altair import ArrowAltairMixin File "/home/user/.local/lib/python3.8/site-packages/streamlit/elements/arrow_altair.py", line 36, in <module> from altair.vegalite.v4.api import Chart ModuleNotFoundError: No module named 'altair.vegalite.v4'
Container logs: |
https://huggingface.co/spaces/IcelandAI/Foods-and-Drinks-of-Iceland | runtime error
Traceback (most recent call last): File "/home/user/.local/bin/streamlit", line 5, in <module> from streamlit.web.cli import main File "/home/user/.local/lib/python3.8/site-packages/streamlit/__init__.py", line 55, in <module> from streamlit.delta_generator import DeltaGenerator as _DeltaGenerator File "/home/user/.local/lib/python3.8/site-packages/streamlit/delta_generator.py", line 45, in <module> from streamlit.elements.arrow_altair import ArrowAltairMixin File "/home/user/.local/lib/python3.8/site-packages/streamlit/elements/arrow_altair.py", line 36, in <module> from altair.vegalite.v4.api import Chart ModuleNotFoundError: No module named 'altair.vegalite.v4'
Container logs: |
https://huggingface.co/spaces/awacke1/acw-dr-llama-7b-chat/tree/main | 1.52 kB initial commit about 1 month ago
272 Bytes Update README.md 20 days ago
32.7 kB Update app.py 1 day ago
30 kB Update backupapp.py 8 days ago
144 Bytes Update requirements.txt about 1 month ago
1.11 kB Create templates.py about 1 month ago |
https://huggingface.co/spaces/awacke1/Health.Assessments.Summarizer | runtime error
Traceback (most recent call last): File "/home/user/.local/bin/streamlit", line 5, in <module> from streamlit.web.cli import main File "/home/user/.local/lib/python3.8/site-packages/streamlit/__init__.py", line 55, in <module> from streamlit.delta_generator import DeltaGenerator as _DeltaGenerator File "/home/user/.local/lib/python3.8/site-packages/streamlit/delta_generator.py", line 45, in <module> from streamlit.elements.arrow_altair import ArrowAltairMixin File "/home/user/.local/lib/python3.8/site-packages/streamlit/elements/arrow_altair.py", line 36, in <module> from altair.vegalite.v4.api import Chart ModuleNotFoundError: No module named 'altair.vegalite.v4'
Container logs: |
https://huggingface.co/spaces/AIZero2HeroBootcamp/3DHuman | App Files Files Community |
https://huggingface.co/spaces/awacke1/PromptSuperHeroImageGenerator/discussions | Welcome to the community
The community tab is the place to discuss and collaborate with the HF community! |
https://huggingface.co/spaces/awacke1/CarePlanQnAWithContext | App Files Files Community |
https://huggingface.co/login?next=%2Fcollections%2Fawacke1%2Fai-favorite-pipelines-650279cfb5d4e364c21adcdf | Log In
Don't have an account? Sign Up
Username or Email address Password
Forgot your password?
SSO is available for companies |
https://huggingface.co/spaces/awacke1/VoiceGPT15 | App Files Files Community |
https://huggingface.co/spaces/AIZero2HeroBootcamp/ChatGPTandLangchain | This Space is sleeping due to inactivity. |
https://huggingface.co/spaces/AIZero2HeroBootcamp/StaticHTML5Playcanvas | App Files Files Community |
https://huggingface.co/spaces/AIZero2HeroBootcamp/AnimatedGifGallery | This Space is sleeping due to inactivity. |
https://huggingface.co/spaces/awacke1/CloneAnyVoice | App Files Files
Community
1 |
https://huggingface.co/spaces/awacke1/ChatGPT-Streamlit-2 | App Files Files |
https://huggingface.co/Classroom-workshop/assignment1-francesco | S2T-SMALL-LIBRISPEECH-ASR
s2t-small-librispeech-asr is a Speech to Text Transformer (S2T) model trained for automatic speech recognition (ASR). The S2T model was proposed in this paper and released in this repository
Model description
S2T is an end-to-end sequence-to-sequence transformer model. It is trained with standard autoregressive cross-entropy loss and generates the transcripts autoregressively.
Intended uses & limitations
This model can be used for end-to-end speech recognition (ASR). See the model hub to look for other S2T checkpoints.
How to use
As this a standard sequence to sequence transformer model, you can use the generate method to generate the transcripts by passing the speech features to the model.
Note: The Speech2TextProcessor object uses torchaudio to extract the filter bank features. Make sure to install the torchaudio package before running this example.
Note: The feature extractor depends on torchaudio and the tokenizer depends on sentencepiece so be sure to install those packages before running the examples.
You could either install those as extra speech dependancies with pip install transformers"[speech, sentencepiece]" or install the packages seperatly with pip install torchaudio sentencepiece.
import torch from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration from datasets import load_dataset model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr") processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr") ds = load_dataset( "patrickvonplaten/librispeech_asr_dummy", "clean", split="validation" ) input_features = processor( ds[0]["audio"]["array"], sampling_rate=16_000, return_tensors="pt" ).input_features # Batch size 1 generated_ids = model.generate(input_ids=input_features) transcription = processor.batch_decode(generated_ids)
Evaluation on LibriSpeech Test
The following script shows how to evaluate this model on the LibriSpeech "clean" and "other" test dataset.
from datasets import load_dataset, load_metric from transformers import Speech2TextForConditionalGeneration, Speech2TextProcessor librispeech_eval = load_dataset("librispeech_asr", "clean", split="test") # change to "other" for other test dataset wer = load_metric("wer") model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr").to("cuda") processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr", do_upper_case=True) librispeech_eval = librispeech_eval.map(map_to_array) def map_to_pred(batch): features = processor(batch["audio"]["array"], sampling_rate=16000, padding=True, return_tensors="pt") input_features = features.input_features.to("cuda") attention_mask = features.attention_mask.to("cuda") gen_tokens = model.generate(input_ids=input_features, attention_mask=attention_mask) batch["transcription"] = processor.batch_decode(gen_tokens, skip_special_tokens=True) return batch result = librispeech_eval.map(map_to_pred, batched=True, batch_size=8, remove_columns=["speech"]) print("WER:", wer(predictions=result["transcription"], references=result["text"]))
Result (WER):
Training data
The S2T-SMALL-LIBRISPEECH-ASR is trained on LibriSpeech ASR Corpus, a dataset consisting of approximately 1000 hours of 16kHz read English speech.
Training procedure
Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization) is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 10,000.
Training
The model is trained with standard autoregressive cross-entropy loss and using SpecAugment. The encoder receives speech features, and the decoder generates the transcripts autoregressively.
BibTeX entry and citation info
@inproceedings{wang2020fairseqs2t, title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq}, author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino}, booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations}, year = {2020}, } |
https://huggingface.co/spaces/AIZero2HeroBootcamp/MultiPDF-QA-ChatGPT-Langchain | This Space is sleeping due to inactivity. |
https://huggingface.co/Classroom-workshop/assignment2-llamas | Edit model card
PPO Agent playing LunarLander-v2
This is a trained model of a PPO agent playing LunarLander-v2 using the stable-baselines3 library.
Usage (with Stable-baselines3)
TODO: Add your code
Evaluation results
mean_reward on LunarLander-v2
self-reported
0.000 |
https://huggingface.co/spaces/awacke1/Image-to-Text-Salesforce-blip-image-captioning-base | App Files Files Community |
https://huggingface.co/spaces/AIZero2HeroBootcamp/VideoToAnimatedGif | Spaces
AIZero2HeroBootcamp
/
VideoToAnimatedGif
Runtime error
App Files Files Community |
https://huggingface.co/Classroom-workshop/assignment1-jack | S2T-SMALL-LIBRISPEECH-ASR
s2t-small-librispeech-asr is a Speech to Text Transformer (S2T) model trained for automatic speech recognition (ASR). The S2T model was proposed in this paper and released in this repository
Model description
S2T is an end-to-end sequence-to-sequence transformer model. It is trained with standard autoregressive cross-entropy loss and generates the transcripts autoregressively.
Intended uses & limitations
This model can be used for end-to-end speech recognition (ASR). See the model hub to look for other S2T checkpoints.
How to use
As this a standard sequence to sequence transformer model, you can use the generate method to generate the transcripts by passing the speech features to the model.
Note: The Speech2TextProcessor object uses torchaudio to extract the filter bank features. Make sure to install the torchaudio package before running this example.
Note: The feature extractor depends on torchaudio and the tokenizer depends on sentencepiece so be sure to install those packages before running the examples.
You could either install those as extra speech dependancies with pip install transformers"[speech, sentencepiece]" or install the packages seperatly with pip install torchaudio sentencepiece.
import torch from transformers import Speech2TextProcessor, Speech2TextForConditionalGeneration from datasets import load_dataset model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr") processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr") ds = load_dataset( "patrickvonplaten/librispeech_asr_dummy", "clean", split="validation" ) input_features = processor( ds[0]["audio"]["array"], sampling_rate=16_000, return_tensors="pt" ).input_features # Batch size 1 generated_ids = model.generate(input_ids=input_features) transcription = processor.batch_decode(generated_ids)
Evaluation on LibriSpeech Test
The following script shows how to evaluate this model on the LibriSpeech "clean" and "other" test dataset.
from datasets import load_dataset, load_metric from transformers import Speech2TextForConditionalGeneration, Speech2TextProcessor librispeech_eval = load_dataset("librispeech_asr", "clean", split="test") # change to "other" for other test dataset wer = load_metric("wer") model = Speech2TextForConditionalGeneration.from_pretrained("facebook/s2t-small-librispeech-asr").to("cuda") processor = Speech2TextProcessor.from_pretrained("facebook/s2t-small-librispeech-asr", do_upper_case=True) librispeech_eval = librispeech_eval.map(map_to_array) def map_to_pred(batch): features = processor(batch["audio"]["array"], sampling_rate=16000, padding=True, return_tensors="pt") input_features = features.input_features.to("cuda") attention_mask = features.attention_mask.to("cuda") gen_tokens = model.generate(input_ids=input_features, attention_mask=attention_mask) batch["transcription"] = processor.batch_decode(gen_tokens, skip_special_tokens=True) return batch result = librispeech_eval.map(map_to_pred, batched=True, batch_size=8, remove_columns=["speech"]) print("WER:", wer(predictions=result["transcription"], references=result["text"]))
Result (WER):
Training data
The S2T-SMALL-LIBRISPEECH-ASR is trained on LibriSpeech ASR Corpus, a dataset consisting of approximately 1000 hours of 16kHz read English speech.
Training procedure
Preprocessing
The speech data is pre-processed by extracting Kaldi-compliant 80-channel log mel-filter bank features automatically from WAV/FLAC audio files via PyKaldi or torchaudio. Further utterance-level CMVN (cepstral mean and variance normalization) is applied to each example.
The texts are lowercased and tokenized using SentencePiece and a vocabulary size of 10,000.
Training
The model is trained with standard autoregressive cross-entropy loss and using SpecAugment. The encoder receives speech features, and the decoder generates the transcripts autoregressively.
BibTeX entry and citation info
@inproceedings{wang2020fairseqs2t, title = {fairseq S2T: Fast Speech-to-Text Modeling with fairseq}, author = {Changhan Wang and Yun Tang and Xutai Ma and Anne Wu and Dmytro Okhonko and Juan Pino}, booktitle = {Proceedings of the 2020 Conference of the Asian Chapter of the Association for Computational Linguistics (AACL): System Demonstrations}, year = {2020}, } |
https://huggingface.co/spaces/awacke1/PromptSuperHeroImageGenerator/tree/main | awacke1
Update README.md
ce68d10 about 1 month ago |
https://huggingface.co/spaces/AIZero2HeroBootcamp/TranscriptAILearnerFromYoutube | This Space is sleeping due to inactivity. |
https://huggingface.co/spaces/AIZero2HeroBootcamp/ClassDescriptionAndExamplesStreamlit | This Space is sleeping due to inactivity. |
https://huggingface.co/spaces/AIZero2HeroBootcamp/ClassDescriptionAndExamples | App Files Files Community |
https://huggingface.co/spaces/AIZero2HeroBootcamp/GetAllContentFromWebURL | runtime error
Scheduling failure: not enough hardware capacity
Container logs:
Fetching error logs... |
https://huggingface.co/cnguye72 | Caitlin Nguyen
cnguye72
Research interests
None yet
Organizations
spaces 1
No application file
🔥
SuperSimple2LinerText2Speech
models
None public yet
datasets
None public yet |
https://huggingface.co/aphn | Andrew Phan
aphn
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/Connorsharp | Connor Sharp
Connorsharp
Research interests
None yet
Organizations
spaces 2
No application file
🔥
SuperSimple2LinerText2Speech
Stopped
🐨
AIPairProgramming1
models
None public yet
datasets
None public yet |
https://huggingface.co/ocrane230249 | Olivia Crane
ocrane230249
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/AliyaJ | J
AliyaJ
Research interests
None yet
Organizations
spaces 1
No application file
🐨
Supersimple2LinerTexttoSpeech
models
None public yet
datasets
None public yet |
https://huggingface.co/gmutombo | Gustave
gmutombo
Research interests
None yet
Organizations
spaces 1
No application file
🐢
SSupersimpletext
models
None public yet
datasets
None public yet |
https://huggingface.co/seanwendlandt | 1 3
Sean Wendlandt
seanwendlandt
Research interests
None yet
Organizations
spaces 5
2
🐢
VideoToAnimatedGif
Stopped
👀
AI Pairprogramming
Runtime error
🐨
SuperSimple2linerText2Speech
🦀
HTML5Example
🏆
HTML5 Interactivity
models
None public yet
datasets
None public yet |
https://huggingface.co/mayamercho | Maya Mercho
mayamercho
Research interests
None yet
Organizations
spaces 1
Stopped
👁
Facebook Fastspeech2 En Ljspeech 0512
models
None public yet
datasets
None public yet |
https://huggingface.co/spaces/Classroom-workshop/assignments-leaderboard | runtime error
Space failed to start. Exit code: 1
Container logs:
Fetching error logs... |
https://huggingface.co/byungchulhan | Byung-Chul Han
byungchulhan
Research interests
None yet
Organizations
spaces 4
🐨
HTML5Example
No application file
💻
SuperSimple2linerText2Speech
🦀
HTML5interactivity
Stopped
👀
AIPairProgramming1
models
None public yet
datasets
None public yet |
https://huggingface.co/Hbali | Harshita Bali
Hbali
Research interests
Big Data Analytics, Deep Learning and Neural Networks, Data Mining, Cluster analysis.
Organizations
spaces 4 |
https://huggingface.co/matt-westerhaus | Matt Westerhaus
matt-westerhaus
mwesterh_uhg
Research interests
None yet
Organizations
spaces 2
🐢
HTML5Interactivity
Runtime error
⚡
AIPairProgramming
models
None public yet
datasets
None public yet |
https://huggingface.co/reecemiron | 1
Reece Miron
reecemiron
Research interests
None yet
Organizations
spaces 2
👀
SuperSimple2LinerTextToSpeech
No application file
👀
AIPairProgramming1
models
None public yet
datasets
None public yet |
https://huggingface.co/willholt | William Holt
willholt
Research interests
None yet
Organizations
spaces 3
Stopped
⛓️
LangFlow
Stopped
1
😻
JAMA GPT
Stopped
📊
AIPairProgramming1
models
None public yet
datasets
None public yet |
https://huggingface.co/dantemoonopt | Dante Moon
dantemoonopt
Research interests
None yet
Organizations
spaces 1
Stopped
🏃
AiPairProgramming1
models
None public yet
datasets
None public yet |
https://huggingface.co/nkelly28 | Nolan Kelly
nkelly28
Research interests
None yet
Organizations
spaces 1
Stopped
👁
SuperSimple2linerText2Speech
models
None public yet
datasets
None public yet |
https://huggingface.co/jzachmann | Jordan Zachmann
jzachmann
Research interests
None yet
Organizations
spaces 5
No application file
📊
VideoToAnimatedGif
🐠
HTML5Example2
No application file
🏆
HTML5Example
Stopped
🏃
SuperSimple2LinerText2Speech
🚀
HTML5Interactivity
models
None public yet
datasets
None public yet |
https://huggingface.co/guysnovelutumba | Guysnove Lutumba
guysnovelutumba
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/misterkait | Kathryn Wright
misterkait
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/m-lubinski13 | Megan Lubinski
m-lubinski13
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/Ccompson | Cameron Compson
Ccompson
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/spaces/AIZero2HeroBootcamp/ExperimentalChatGPTv1 | This Space is sleeping due to inactivity. |
https://huggingface.co/madelinekirke | Madeline Kirke
madelinekirke
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/ninapd | Nina Perez-Dubson
ninapd
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/garrettv | Garrett Vareldzis
garrettv
Research interests
None yet
Organizations
spaces 3
No application file
📊
HTML5interactivityDemo
Runtime error
📉
Facebook Fastspeech2 En Ljspeech 1111
No application file
🐨
Facebook Fastspeech2 En Ljspeech Gv
models
None public yet
datasets
None public yet |
https://huggingface.co/lainestubbs | Laine Stubbs
lainestubbs
Research interests
None yet
Organizations
spaces 1
Stopped
🏃
Facebook Fastspeech2 En Ljspeech 0731
models
None public yet
datasets
None public yet |
https://huggingface.co/samschock | Samuel Schock
samschock
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/lekreitzer | Lauren Kreitzer
lekreitzer
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/jesterkq | Keely Jester
jesterkq
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/spaces/awacke1/WikipediaProfilerTestforDatasets | App Files Files Community |
https://huggingface.co/hannahross5 | 3
Hannah Ross
hannahross5
Research interests
None yet
Organizations
spaces 3
🐠
HTML5interactivitydemo
Stopped
1
📚
Memory
Runtime error
1
⚡
Facebook Fastspeech2 En Ljspeech 0731
models
None public yet
datasets
None public yet |
https://huggingface.co/adianaleidich | Adiana Leidich
adianaleidich
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/jadaiacco | Jada Iacco
jadaiacco
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/sydneyharper | Sydney Harper
sydneyharper
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/jsamuelson | John Samuelson
jsamuelson
Research interests
None yet
Organizations
spaces 1
Stopped
📈
Facebook Fastspeech2 En Ljspeech 069710
models
None public yet
datasets
None public yet |
https://huggingface.co/kyliefrese | Kylie Frese
kyliefrese
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/kaptainkadar | Alex Kadar
kaptainkadar
Research interests
None yet
Organizations
spaces 2
🏆
HTML5interactivitydemo
Stopped
🌍
Facebook Fastspeech2 En Ljspeech Kk0731
models
None public yet
datasets
None public yet |
https://huggingface.co/dfinck1 | Daniel Casey Finck
dfinck1
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/lcunningham22 | Alexis Cunningham
lcunningham22
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/Nidian | elizondo
Nidian
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/spaces/Yntec/ToyWorld | App Files Files
Community
3 |
https://huggingface.co/bbaker272514 | Bridget Baker
bbaker272514
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/brettstrandskov | Brett Strandskov
brettstrandskov
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/anyarob | Anya Roberts
anyarob
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/mlito17 | Mario Lito
mlito17
Research interests
None yet
Organizations
spaces 2
Stopped
📊
SuperSimpleTwoLineText2Speech
Stopped
🐨
AIPairProgramming1
models
None public yet
datasets
None public yet |
https://huggingface.co/sofiapostigo | Sofia Postigo
sofiapostigo
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/msalcena | Mathew Salcena
msalcena
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/sawyersmith1 | 2
Sawyer Smith
sawyersmith1
Research interests
None yet
Organizations
spaces 2
🐠
HTML5DemoActivity
Stopped
🐨
Facebook Fastspeech2 En Ljspeech 0998
models
None public yet
datasets
None public yet |
https://huggingface.co/darbysween | Darby Sween
darbysween
Research interests
None yet
Organizations
spaces 1
Stopped
😻
ChatGPTandLangchain
models
None public yet
datasets
None public yet |
https://huggingface.co/josiegreener | josephine greener
josiegreener
Research interests
None yet
Organizations
spaces 1
No application file
📈
HTML5demo
models
None public yet
datasets
None public yet |
https://huggingface.co/zfrieseke | Zachary Frieseke
zfrieseke
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/willdeters2 | William Deters
willdeters2
Research interests
None yet
Organizations
models
None public yet
datasets
None public yet |
https://huggingface.co/login?next=%2FAIZero2HeroBootcamp | Log In
Don't have an account? Sign Up
Username or Email address Password
Forgot your password?
SSO is available for companies |
Subsets and Splits