Seed ASR 2.0 — Reelina AI साउंड
Audio container format.
Optional language code. Leave empty to auto-recognize Chinese/English/dialects. Or set a code, e.g. en-US, zh-CN, yue-CN, ja-JP, es-MX (51 languages supported).
enable itn
Inverse text normalization (e.g. 'one hundred' -> '100').
enable punc
Add punctuation to the transcript.
enable ddc
Semantic smoothing (remove fillers/repeats).
enable speaker info
Speaker diarization (<=10 speakers).
show utterances
Return per-sentence segments with timestamps.
About Seed ASR 2.0
BytePlus Seed Speech recording-file recognition (ASR 2.0): transcribe an audio file to text with punctuation, ITN, sentence segmentation, speaker diarization, and 51-language support.
Specifications
| Provider | ByteDance |
|---|---|
| Type | Audio generation |
| Cost per run | 1 credits |
| Input | Text prompt |
Alternatives to Seed ASR 2.0
Models in this workspace that do the same job, closest in price first. Your credit balance works across all of them.
Frequently asked questions
- What is Seed ASR 2.0?
- BytePlus Seed Speech recording-file recognition (ASR 2.0): transcribe an audio file to text with punctuation, ITN, sentence segmentation, speaker diarization, and 51-language support.
- How much does Seed ASR 2.0 cost?
- Seed ASR 2.0 costs 1 credits per run, deducted from your Reelina balance. The exact cost is shown before you press Generate, and one balance covers every model — there is no separate subscription per model.
- What do I need to use Seed ASR 2.0?
- A text prompt describing what you want. No upload needed.
- Can I use the results commercially?
- Your generations are yours to use, subject to our terms and acceptable use policy. For anything business-critical, check the licence of the specific model you pick.
- How long are my files kept?
- अपलोड सिर्फ़ आपका रिज़ल्ट तैयार करने के लिए इस्तेमाल किए जाते हैं। क्रिएशन डिफ़ॉल्ट रूप से प्राइवेट होते हैं और जनरेट होने के एक हफ्ते बाद सब कुछ डिलीट कर दिया जाता है।
Looking for something else? Browse every model.