Saltar al contenido
ModelPublicadoExterno
Descargar

Swecha Telangana Asr Base Model

The Swecha Telangana ASR Base Model is a fine-tuned Automatic Speech Recognition (ASR) model built on OpenAI Whisper Small for Telugu speech recognition.

Fuentehuggingface.co/akhila56/swecha-telangana-asr-base-model/tree/09c7faab72a9dafce9ae966f10a2f3f7be46d622
TipoText modelModelo baseopenai/whisper-smallVersión

Fuente del modelo

Tipo
Text model
Tamaño del archivo
922 MB
Modelo base
openai/whisper-small
Versión
vmodel
Actualizado
22 sept 2026
Descargas
7
Fuente
Hugging Face

Descripción de la fuente

Swecha Telangana ASR Base Model

Model Description

The Swecha Telangana ASR Base Model is a fine-tuned Automatic Speech Recognition (ASR) model built on OpenAI Whisper Small for Telugu speech recognition.

This model has been fine-tuned using the Mozilla Common Voice Scripted Speech 26.0 Telugu dataset through the Mozilla AI Speech-to-Text Finetuning Blueprint.

Base Model

  • openai/whisper-small

Language

  • Telugu (te)

Training Dataset

  • Mozilla Common Voice Scripted Speech 26.0 – Telugu
  • License: CC0-1.0

Training Configuration

ParameterValue
Base ModelWhisper Small
LanguageTelugu
Learning Rate1e-5
Batch Size32
Gradient Accumulation Steps1
Warmup Steps50
Max Steps50
Mixed PrecisionFP16
Evaluation MetricWord Error Rate (WER)

Intended Uses

  • Telugu Speech Recognition
  • Speech-to-Text Applications
  • Voice Assistants
  • Accessibility Tools
  • Educational Applications
  • Automatic Caption Generation
  • Research in Indian Language ASR

Limitations

  • Trained only on Mozilla Common Voice Telugu speech.
  • Performance may decrease for:
    • Noisy environments
    • Code-mixed speech
    • Unseen accents
    • Domain-specific vocabulary

Evaluation Results

MetricBefore Fine-tuningAfter Fine-tuning
WER211.7175.92
CER174.0246.38

The fine-tuned model significantly improves Telugu speech recognition compared to the baseline Whisper Small model.

Citation

If you use this model, please cite:

  • Mozilla Common Voice
  • Mozilla AI Speech-to-Text Finetuning Blueprint
  • OpenAI Whisper

Author

Akhila Vishwanath

Hugging Face: https://huggingface.co/akhila56

Fuentes

1 fuente
Hugging FacePrincipalVersión 09c7faab72a9dafce9ae966f10a2f3f7be46d622
huggingface.co/akhila56/swecha-telangana-asr-base-model/tree/09c7faab72a9dafce9ae966f10a2f3f7be46d622

Verificado 22 sept

Artefactos del modelo

2 artefactos

checkpoint-50/model.safetensors

safetensors · 922 MB · SHA-256 40b2fc86ab33…11ac · Hugging Face

Descargar

model.safetensors

safetensors · 922 MB · SHA-256 40b2fc86ab33…11ac · Hugging Face

Descargar

Extractos de fuentes

2 extractos
Readme22 septhuggingface.co/akhila56/swecha-telangana-asr-base-model

--- language: - te license: apache-2.0 base_model: openai/whisper-small pipeline_tag: automatic-speech-recognition tags: - automatic-speech-recognition - speech-to-text - whisper - telugu - asr - swecha - telangana --- # Swecha Telangana ASR Base Model ## Model Description The **Swecha Telangana ASR Base Model** is a fine-tuned Automatic Speech Recognition (ASR) model built on **OpenAI Whisper Small** for Telugu speech recognition. This model has been fine-tuned using the **Mozilla Common Voice Scripted Speech 26.0 Telugu** dataset through the Mozilla AI Speech-to-Text Finetuning Blueprint. ## Base Model - openai/whisper-small ## Language - Telugu (te) ## Training Dataset - Mozilla Common Voice Scripted Speech 26.0 – Telugu - License: CC0-1.0 ## Training Configuration | Parameter | Value | |-----------|-------| | Base Model | Whisper Small | | Language | Telugu | | Learning Rate | 1e-5 | | Batch Size | 32 | | Gradient Accumulation Steps | 1 | | Warmup Steps | 50 | | Max Steps | 50 | | Mixed Precision | FP16 | | Evaluation Metric | Word Error Rate (WER) | ## Intended Uses - Telugu Speech Recognition - Speech-to-Text Applications - Voice Assistants - Accessibility Tools - Educational Applications - Automatic Caption Generation - Research in Indian Language ASR ## Limitations - Trained only on Mozilla Common Voice Telugu speech. - Performance may decrease for: - Noisy environments - Code-mixed speech - Unseen accents - Domain-specific vocabulary ## Evaluation Results | Metric | Before Fine-tuning | After Fine-tuning | |--------|--------------------:|------------------:| | WER | 211.71 | 75.92 | | CER | 174.02 | 46.38 | The fine-tuned model significantly improves Telugu speech recognition compared to the baseline Whisper Small model. ## Citation If you use this model, please cite: - Mozilla Common Voice - Mozilla AI Speech-to-Text Finetuning Blueprint - OpenAI Whisper ## Author **Akhila Vishwanath** Hugging Face: https://huggingface.co/akhila56

Note22 septhuggingface.co/akhila56/swecha-telangana-asr-base-model

Source context: 7 downloads · 0 likes · Pipeline automatic-speech-recognition · Repo akhila56/swecha-telangana-asr-base-model