seed-asr-bigmodel is ByteDance's Seed-ASR large-model speech recognition service on BytePlus, exposed through the audio-file recognition API as the bigmodel engine.
- Input
- audio
- Output
- text
- Price
- $0.002/min
Price in context
Where the price sits among 11 comparable models
The bar shows how this model’s price compares with every other model of the same kind on Synthorai. The cheapest and the most expensive are named at each end. These are base rates; batch, region and cache-write discounts are on the pricing page.
Specs & limits
Audio
| Languages | With `language` empty the model covers Mandarin, English, Cantonese, Shanghainese, Minnan, Sichuan and Shaanxi dialects; 39 language keys can be pinned explicitly (en-US, zh-CN, yue-CN, ja-JP, ko-KR, id-ID, es-MX, pt-BR, de-DE, fr-FR, fil-PH, ms-MY, th-TH, ar-SA, it-IT, bn-BD, el-GR, nl-NL, ru-RU, tr-TR, vi-VN, pl-PL, ro-RO, uk-UA, az-AZ, bg-BG, cs-CZ, da-DK, fi-FI, hi-IN, hu-HU, kk-KZ, km-KH, my-MM, no-NO, pa-PK, sv-SE, sw-KE, ur-PK), plus optional auto-detection (enable_auto_lang) |
|---|---|
| Audio limits |
|
| Speaker diarization | Yes |
| Streaming transcription | Yes |
| Timestamps | Yes |
Model
| Modalities | audio → text |
|---|
No official model id 'seed-asr-bigmodel': BytePlus Seed Speech uses model_name 'bigmodel' with resource ids volc.bigasr.auc (Seed ASR 1.0) / volc.seedasr.auc (Seed ASR 2.0).
Use Seed ASR in 30 seconds
OpenAI-compatible: swap the base_url, keep your SDK. POST /v1/audio/transcriptions
from openai import OpenAI
client = OpenAI(
base_url="https://synthorai.io/v1",
api_key="sk-syn-...",
)
resp = client.audio.transcriptions.create(
model="seed-asr-bigmodel",
file=open("meeting.mp3", "rb"),
language="en",
)
print(resp.text)import OpenAI from "openai";
import fs from "node:fs";
const client = new OpenAI({
baseURL: "https://synthorai.io/v1",
apiKey: "sk-syn-...",
});
const resp = await client.audio.transcriptions.create({
model: "seed-asr-bigmodel",
file: fs.createReadStream("meeting.mp3"),
});
console.log(resp.text);curl https://synthorai.io/v1/audio/transcriptions \
-H "Authorization: Bearer sk-syn-..." \
-F model="seed-asr-bigmodel" \
-F file=@meeting.mp3package main
import (
"context"
"fmt"
"os"
"github.com/openai/openai-go/v3"
"github.com/openai/openai-go/v3/option"
)
func main() {
client := openai.NewClient(
option.WithBaseURL("https://synthorai.io/v1"),
option.WithAPIKey("sk-syn-..."),
)
f, _ := os.Open("meeting.mp3")
resp, _ := client.Audio.Transcriptions.New(context.TODO(), openai.AudioTranscriptionNewParams{
Model: "seed-asr-bigmodel",
File: f,
})
fmt.Println(resp.Text)
}import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.audio.transcriptions.*;
import java.nio.file.Paths;
OpenAIClient client = OpenAIOkHttpClient.builder()
.baseUrl("https://synthorai.io/v1")
.apiKey("sk-syn-...")
.build();
Transcription resp = client.audio().transcriptions().create(
TranscriptionCreateParams.builder()
.model("seed-asr-bigmodel")
.file(Paths.get("meeting.mp3"))
.build()).asTranscription();
System.out.println(resp.text());About Seed ASR
- It transcribes recordings asynchronously via a submit-and-query flow suited to batch and offline workloads: the caller supplies its own request ID, which doubles as the task ID, then polls until the status code reports completion, with results produced within three hours and retrievable for seven days.
- Audio can run up to five hours and 512 MB across OPUS, WAV, MP3, OGG, AMR, AAC, and M4A, in mono or stereo.
- It recognizes Mandarin, English, Cantonese, and several Chinese dialects by default, with dozens of settable languages from Japanese to Arabic and optional automatic language detection, which takes precedence over an explicitly named language when both are set.
- Hotword biasing and conversational context improve accuracy, with hotword lists capped at 5,000 words and dialog context at 800 tokens or twenty rounds, and Seed ASR 2.0 extends that context to an accompanying image, one per request.
- Speaker diarization, channel splitting, utterance and word-level timestamps with confidence, sensitive-word filtering, and traditional Chinese variants are all optional flags; note that punctuation is off by default while inverse text normalization is on, and that diarization is a file-mode feature the streaming modes do not offer.
- Synthorai wraps it in an OpenAI-compatible transcription endpoint.
FAQ
Is the Seed ASR API free to try?
Yes: new accounts get 10 trial calls and up to $1 in free credit, no card required. That's enough to try Seed ASR against your real workload before adding a payment method.
What is Seed ASR best at?
Asynchronous submit-and-query batch transcription; dozens of settable languages with auto-detection; conversational context extends to an accompanying image. See the About section for the full picture from the vendor's own release notes.
How much does Seed ASR cost?
Seed ASR costs $0.002 per minute of audio transcribed on Synthorai: pay-as-you-go, no platform markup, no subscription.
Which languages does Seed ASR support?
Seed ASR supports With `language` empty the model covers Mandarin, English, Cantonese, Shanghainese, Minnan, Sichuan and Shaanxi dialects; 39 language keys can be pinned explicitly (en-US, zh-CN, yue-CN, ja-JP, ko-KR, id-ID, es-MX, pt-BR, de-DE, fr-FR, fil-PH, ms-MY, th-TH, ar-SA, it-IT, bn-BD, el-GR, nl-NL, ru-RU, tr-TR, vi-VN, pl-PL, ro-RO, uk-UA, az-AZ, bg-BG, cs-CZ, da-DK, fi-FI, hi-IN, hu-HU, kk-KZ, km-KH, my-MM, no-NO, pa-PK, sv-SE, sw-KE, ur-PK), plus optional auto-detection (enable_auto_lang). On Synthorai you call it through POST /v1/audio/transcriptions, the OpenAI transcription API shape.
How do I get access to Seed ASR?
Point your existing OpenAI SDK at base_url="https://synthorai.io/v1", set model="seed-asr-bigmodel", and you're done. One API key covers every model on the gateway.
Related models
Compare
Every value on this page is transcribed from the vendor's own documentation, linked above, and carries the date it was checked. Prices are compared across the catalogue; specification values that vendors define differently are shown with the difference stated rather than charted. Nothing here is measured by us, and nothing is scored.