Vikhrmodels
/

salt-asr_wav-uni_1_tts_wav-uni_1-12k

Safetensors

llama

Model card Files Files and versions Community

AlexWortega commited on Jan 23

Commit

c2d8e8d

verified ·

1 Parent(s): 82ac092

Update README.md

Browse files

Files changed (1) hide show

README.md +54 -179

README.md CHANGED Viewed

@@ -1,199 +1,74 @@
----
-library_name: transformers
-tags: []
----
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
-## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-### Model Sources [optional]
-<!-- Provide the basic links for the model. -->
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-## Uses
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-### Direct Use
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-[More Information Needed]
-### Downstream Use [optional]
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-[More Information Needed]
-### Out-of-Scope Use
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-[More Information Needed]
-## Bias, Risks, and Limitations
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-[More Information Needed]
-### Recommendations
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-## How to Get Started with the Model
-Use the code below to get started with the model.
-[More Information Needed]
-## Training Details
-### Training Data
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-[More Information Needed]
-### Training Procedure
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-#### Preprocessing [optional]
-[More Information Needed]
-#### Training Hyperparameters
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-#### Speeds, Sizes, Times [optional]
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-[More Information Needed]
-## Evaluation
-<!-- This section describes the evaluation protocols and provides the results. -->
-### Testing Data, Factors & Metrics
-#### Testing Data
-<!-- This should link to a Dataset Card if possible. -->
-[More Information Needed]
-#### Factors
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-[More Information Needed]
-#### Metrics
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-[More Information Needed]
-### Results
-[More Information Needed]
-#### Summary
-## Model Examination [optional]
-<!-- Relevant interpretability work for the model goes here -->
-[More Information Needed]
-## Environmental Impact
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
-- **Hardware Type:** [More Information Needed]
-- **Hours used:** [More Information Needed]
-- **Cloud Provider:** [More Information Needed]
-- **Compute Region:** [More Information Needed]
-- **Carbon Emitted:** [More Information Needed]
-## Technical Specifications [optional]
-### Model Architecture and Objective
-[More Information Needed]
-### Compute Infrastructure
-[More Information Needed]
-#### Hardware
-[More Information Needed]
-#### Software
-[More Information Needed]
-## Citation [optional]
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-**BibTeX:**
-[More Information Needed]
-**APA:**
-[More Information Needed]
-## Glossary [optional]
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-[More Information Needed]
-## More Information [optional]
-[More Information Needed]
-## Model Card Authors [optional]
-[More Information Needed]
-## Model Card Contact
-[More Information Needed]

+### English Version 🇬🇧
+---
+#### **Model Performance Overview**
+**Metrics**:
+- **PESQ@200**: Perceptual Evaluation of Speech Quality (higher = better).
+- **STOI@200**: Short-Time Objective Intelligibility (closer to 1 = better).
+- **SI-SDR@200**: Scale-Invariant Signal-to-Distortion Ratio (higher = better).
+- **SIM-O@200**: Similarity to ground truth (higher = better).
+| Model                     | PESQ@200       | STOI@200      | SI-SDR@200        | SIM-O@200      |
+|---------------------------|----------------|---------------|-------------------|----------------|
+| Original (LibriSpeech)    | 4.15           | 0.997         | 27.45 ±1.09       | —              |
+| Parler TTS Mini v1        | 1.29 ±0.49     | 0.15 ±0.12    | 25.0 ±2.9         | 0.88 ±0.03     |
+| Fish Speech 1.5           | 1.26 ±0.38     | 0.17 ±0.12    | 25.0 ±3.2         | 0.91 ±0.02     |
+| **Salt-ASR Wav-Uni 1-12k **  | **1.27 ±0.40**     | 0.18 ±0.09    | 20.3 ±3.69        | 0.88 ±0.02     |
+---
+#### **Our Solution**
+- **Method**: Extends a pre-trained LLM with audio tokens and fine-tunes on **TTS** and **ASR** tasks.
+- **Training**:
+  - SpeechTokenizer (semantic + audio tokens) outperformed Encodec (loss explosions resolved with TF32 precision).
+  - Training time: **150 A100 GPU hours**.
+- **Advantages**: Unified LM loss for dual tasks, minimal training overhead.
+---
+#### **Resources**
+- Code: [GitHub Repo](https://github.com/VikhrModels/Vikhr4o)
+- Inference Demo: [Google Colab](https://colab.research.google.com/drive/1Poz6jNJu7-HRIkRkPVTzEqjJ2qKn4eUt)
+- Reference Papers: [Vitta](https://arxiv.org/pdf/2408.05211), [Valle](https://github.com/lifeiteng/vall-e)
+---
+### Русская Версия 🇷🇺
+---
+#### **Сравнение моделей**
+**Метрики**:
+- **PESQ@200**: Качество речи (чем выше, тем лучше).
+- **STOI@200**: Разборчивость речи (ближе к 1 = лучше).
+- **SI-SDR@200**: Соотношение сигнал-шум (выше = лучше).
+- **SIM-O@200**: Сходство с эталоном (выше = лучше).
+| Модель                   | PESQ@200       | STOI@200      | SI-SDR@200        | SIM-O@200      |
+|--------------------------|----------------|---------------|-------------------|----------------|
+| Original (LibriSpeech)   | 4.15           | 0.997         | 27.45 ±1.09       | —              |
+| Parler TTS Mini v1       | 1.25 ±0.49     | 0.15 ±0.12    | 25.0 ±2.9         | 0.88 ±0.03     |
+| Fish Speech 1.5          | 1.26 ±0.38     | 0.17 ±0.12    | 25.0 ±3.2         | 0.91 ±0.02     |
+| **Salt-ASR Wav-Uni 1-12k **  | **1.27 ±0.40**     | 0.18 ±0.09    | 20.3 ±3.69        | 0.88 ±0.02     |
+---
+#### **Наше решение**
+- **Метод**: Расширение словаря LLM аудиотокенами + дообучение на **TTS** и **ASR**.
+- **Обучение**:
+  - SpeechTokenizer (семитические + аудиотокены) показал лучшие результаты, чем Encodec.
+  - Время обучения: **150 часов на A100**.
+- **Преимущества**: Единая функция потерь для двух задач, минимальные затраты.
+---
+#### **Ресурсы**
+- Код: [GitHub](https://github.com/VikhrModels/Vikhr4o)
+- Демо: [Google Colab](https://colab.research.google.com/drive/1Poz6jNJu7-HRIkRkPVTzEqjJ2qKn4eUt)
+---
+**Примечание**: Модель поддерживает генерацию коротких фраз на английском, немецком и французском.