inesc-id/camoes_asr

CAMÕES

CAMÕES is a comprehensive and diverse evaluation benchmark for European Portuguese (EP) Automatic Speech Recognition (ASR).

The benchmark is organized into five distinct domains, enumerated below:

• Read Speech (RS): Read audiobooks and text prompts, such as news articles, speech commands, numbers, single words and digits, with little to no spontaneity.
• Broadcast News (BN): News content from public Portuguese TV channels, chosen as an individual domain due to its particularities, i.e., mostly read or planned speech with a specific type of enunciation, uttered by professionals.
• Talks/Lectures (T/L): TEDx talks and university lectures; this type of speech is prepared but not read, with a higher degree of spontaneity than previous domains.
• Conversational Speech (CS): Celebrity interviews, map task dialogues, and other recordings from Portuguese TV channels.
• Sociolinguistic Interviews (SI): Highly spontaneous conversational speech recorded in various Portuguese regions and social contexts.

This repository does not contain audio data. It serves as a central hub for information about the benchmark, including the leaderboard. We invite the community to submit their models for evaluation.

Leaderboard

Rank	Model	# Trainable Params	RS	BN	T/L	CS	SI	Avg. WER
1	WhisperLv3-X (FT)	1.55B	7.4	4.7	11.3	11.2	27.9	12.5
2	WhisperLv3-X (PT-All)	1.55B	7.9	4.7	12.3	11.6	28.6	13.0
3	WhisperLv3-X (AAP)	1.55B	8.2	5.9	12.1	11.7	28.9	13.4
4	EBranch-w2vBERT2 + 4-gram LM (TFS)	114M	8.0	5.4	15.6	13.4	27.1	13.9
5	EBranch-w2vBERT2 (AAP)	114M	8.3	5.5	16.0	13.7	27.4	14.2
6	EBranch-w2vBERT2 (TFS)	114M	8.3	5.4	16.0	14.9	27.2	14.4
6	EBranch-w2vBERT2 (PT-All)	114M	8.7	5.5	16.4	13.5	28.0	14.4
8	EBranch-XLSR (TFS)	114M	9.6	6.5	16.7	18.2	29.3	16.1
9	WhisperLv3 (FT)	1.55B	7.2	4.6	13.6	14.9	43.2	16.7
10	EBranch (TFS)	114M	9.4	6.5	18.0	16.5	35.4	17.2
11	WhisperLv3-X (0-shot)	-	16.4	8.2	16.6	15.3	39.3	19.2
12	Phi-4-MI (0-shot)	-	15.5	8.6	17.9	21.9	44.5	21.7
13	Phi-4-MI (FT)	1.3B	9.6	7.2	16.7	24.4	59.5	23.5
14	WhisperLv3-X (BP)	1.55B	17.2	13.8	24.1	20.8	46.6	24.5
15	WhisperLv3 (0-shot)	-	32.4	7.9	15.4	18.3	49.0	24.6
16	OWSM-CTC v4 (0-shot)	-	22.5	24.4	32.0	28.7	52.1	31.9
17	SeamlessM4T-v2 (0-shot)	-	26.7	17.3	26.3	27.9	64.5	32.5
18	MMS-all (0-shot)	-	33.9	25.7	40.3	38.2	65.4	40.7
19	EBranch-w2vBERT2 (BP)	114M	37.9	32.6	42.2	40.0	54.9	41.5

How to Participate

We invite the community to submit their ASR models for evaluation on the CAMÕES benchmark using the following portal:

Official Submission Portal

Model Requirements

The submitted model must be uplodead to Hugging face and be loadable and usable with the pipeline("automatic-speech-recognition") function as such:

from transformers import pipeline

# Load your model from the Hugging Face Hub
asr_pipeline = pipeline("automatic-speech-recognition", model="model_path")

# Run inference
result = asr_pipeline("path/to/audio.wav")

Benchmark Details

Benchmark Sources

Domain	Corpus	Test (Hrs)	Test (#Spks)	Age	M|F (%)	Notes
RS	BD-Publico	2.0	10	18-28	50|50	Read sentences extracted from an EP newspaper.
	CommonVoice	1.8	42	13-59	48|12	Speaker count estimated from the client ids provided in the corpus.
	MLS_extended	1.0	10	NI	27|73	EP extension of MLS: automatically aligned audiobooks.
	PT_Adults	1.6	17	25-59	52|48	Corresponds to YMA.
	PT_Children	2.1	52	3-10	56|44	Corpus of child speech.
	PT_Elderly	1.3	172	60-100	26|74	Speakers are aged between 76-100 years.
	SpeechDat	9.7	604	14-98	46|54	Telephone speech sampled at 8kHz, upsampled to 16kHz.
BN	Alert	6.6	175	NI	70|29	Broadcast news data.
T/L	Lectra	2.6	7	NI	NI	University lectures. Speakers are shared among partitions.
	MuAViC	0.4	2	NI	60|40	TEDx talks.
CS	Postport	3.9	>30	NI	54|24	Debates and entertainment.
	VoxCelebPT	2.9	13	NI	38|62	Voices of Portuguese celebrities collected from YouTube.
SI	Fala Bracarense	6.1	8	15-92	45|55	Recorded in the city of Braga, collected between 2009-2014.
	PT Fundamental	4.2	169	17-69	44|56	Low quality recordings of interviews collected in the 1970’s.
Total		46.2	1,311

For additional details regarding the benchmark and corresponding baselines please refer to the paper.

Citation

BibTeX:

@inproceedings{camoes,
    title={{CAMÕES: A Comprehensive Automatic Speech Recognition Benchmark for  European Portuguese}},
    author={Carlos Carvalho, Francisco Teixeira, Catarina Botelho, Anna Pompili, Rubén Solera-Ureña, Sérgio Paulo, Mariana Julião, Thomas Rolland, John Mendonça, Diogo Pereira, Isabel Trancoso, Alberto Abad},
    booktitle={Proceedings of the IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)},
    year={2025},
}

• Curated by: Human Technologies Lab INESC-ID Lisboa
• Funded by: Portuguese national funds through Fundação para a Ciência e a Tecnologia, under project UIDB/50021/2020 (DOI:10.54499/UIDB/50021/2020), by the Portuguese Recovery and Resilience Plan and NextGenerationEU European Union funds under project C644865762-00000008 (Accelerat.AI).
• License: Other