Syrix Labelling - Personal Dataset

Collection:

🎤

Collection Name

general

Topic

No topic specified

Description

No description available

Recording Guidelines

No guidelines provided

Collection Progress

Hours Recorded 0 / 0

Unique Speakers 0 / 0

0 Segments

0s Avg/Segment

0 / 0 STT Done

0

Unique Speakers

0

Total Embeddings

75%

Similarity Threshold

Processing...

Speaker Clusters

Speaker ID	Appearances	Audio Files
No data available

Analysis completed at: --

User	Total	Validated	Rejected	Pending	Val. Rate	Avg Quality
Loading... Loading statistics...

Choose your preferred transcription engine and configure settings

Transcription Engine

ONNX (Local ASR) FREE FAST

Loading model info... - Local inference, no cost, optimized for Vietnamese

Google Gemini AI PAID CLOUD

Cloud-based AI with multilingual support and high accuracy

Checking system API key...

Your Gemini API Key Get FREE key

Gemini Model Select based on your needs: Flash models are faster & cheaper, Pro models are more accurate

Pricing Info:

~$0.01-0.05 per minute of audio (varies by model)
Free tier: 15 requests/minute, 1500/day
Get your free API key from Google AI Studio

Audio Enhancement

Enable Audio Enhancement
Reduces noise, echo, and improves clarity. Best for poor quality audio (meetings, outdoor recordings). May reduce accuracy for clean audio.

Recommendation: Use ONNX for Vietnamese audio (free, fast). Use Gemini for multilingual content or when you need higher accuracy.

Register New Voice

Audio File WAV/MP3 preferred, duration < 10s

Transcription (7–10 words) Words: 0/10

Name

Age

Gender

Province

Region

Accent

Voice Type

My Ref Voices

Statistics

By Region

—

By Gender

—

By Status

—

Total Duration

—

My TTS Prompt List Assignments

No TTS prompt lists assigned to you yet

Admin will assign TTS prompt lists from the admin dashboard

- -

REF VOICE

Type

SPEED: 0.8x

0.5x 1.5x

GENERATE

From prompt #

To prompt #

Clear from #

to #

AI RECHECK

Click ⚙️ to configure models

From prompt #

To prompt #

Will recheck: 1 → end

CURRENT PROMPT

- Select a prompt →

AUDIO -

ORIGINAL TEXT

Select a prompt from the right →

TRANSCRIPT

PROMPTS

○ 0 Pending ◐ 0 Processing ● 0 AI Check ● 0 Recheck ● 0 Done ✕ 0 Error

Total Audio 0s

Payment

What you have earned, and what has been transferred.

Pending

— ₫

Estimated, not yet transferred

Received

— ₫

All transfers to date

This month

— ₫

Transferred since the 1st

My work —

Task type

Status

Admin review

Payment

Item	Payment	Type	Collection	Status	Admin review	Progress	Assigned	Completed	Hours	Rate / hour	Estimated	Paid
Loading…

Payments —

Status

Month

Payment	Period	Created	Paid on	Tasks	Hours	Amount	Status	Method	Proof
Loading…

Select one ASR model for recheck analysis

Syrix ASR FREE LOCAL

Local ONNX model, Vietnamese-optimized, no cost

Model & Epoch

Batch Size (recommended: 30)

Parallel Workers (recommended: 8)

Google Gemini AI PAID CLOUD

Multilingual cloud model, high accuracy

Gemini API Key

Model

Alibaba Qwen3 FREE LOCAL

Local Qwen3-ASR model, 30 languages, runs on GPU

Language

Batch Size (recommended: 20)

Parallel Workers (recommended: 4)

✅ Qwen3-ASR-1.7B — supports 30 languages, auto language detection

VibeVoice COMING SOON

Advanced multilingual ASR (not yet available)

Collection Name

My Work Assignments

No tasks assigned yet

Audio Enhancement

Upload Audio File

Upload Settings

Register New Voice

My Ref Voices

Statistics

My TTS Prompt List Assignments

Payment

Collection Name

My Work Assignments

No tasks assigned yet

Audio Enhancement

Upload Audio File

Upload Settings

Dataset Statistics

Recent Activity

Speaker Analysis Results

0

0

75%

Speaker Clusters

Add Speaker

Merge Segments

Split Segment

Adjust Boundaries

Provide Feedback

User Statistics

Enhanced AI Transcription Settings

AI Transcription Settings

Register New Voice

My Ref Voices

Statistics

My TTS Prompt List Assignments

Payment

Transfer proof

Confirm audio deletion

AI Recheck Settings