Paraformer-Realtime-v2 is Alibaba Cloud's earlier-generation real-time speech recognition model, streaming audio in and transcripts out over a bidirectional connection.
- 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 | 7 languages: Chinese (Mandarin plus Cantonese, Wu, Hokkien and 15 regional accents), English, Japanese, Korean, German, French, Russian |
|---|---|
| Audio limits |
|
| Streaming transcription | Yes |
| Timestamps | Yes |
Model
| Modalities | audio → text |
|---|
Streaming-first Paraformer generation with hotword customization and multilingual coverage.
Use Paraformer Realtime v2 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="paraformer-realtime-v2",
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: "paraformer-realtime-v2",
file: fs.createReadStream("meeting.mp3"),
});
console.log(resp.text);curl https://synthorai.io/v1/audio/transcriptions \
-H "Authorization: Bearer sk-syn-..." \
-F model="paraformer-realtime-v2" \
-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: "paraformer-realtime-v2",
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("paraformer-realtime-v2")
.file(Paths.get("meeting.mp3"))
.build()).asTranscription();
System.out.println(resp.text());About Paraformer Realtime v2
- It recognizes Chinese Mandarin along with an unusually wide set of regional dialects (Cantonese, Wu, Hokkien, Sichuan, and more) plus English, Japanese, Korean, German, French, and Russian, and supports a language_hints parameter for guiding detection.
- That is a v2-only addition, as is acceptance of arbitrary sample rates where the previous version required 16 kHz.
- Alibaba's docs now recommend Fun-ASR or Qwen-ASR for new projects, positioning this model for existing integrations.
- What keeps it usable is a mature parameter surface: hotword tables for domain vocabulary, inverse text normalization on by default, punctuation prediction on by default with optional semantic punctuation, a configurable maximum sentence silence that closes an utterance, and an optional disfluency filter that strips filler words (a control the Fun-ASR realtime model does not document).
- Sentence- and word-level timestamps are returned in milliseconds, and the connection accepts PCM, WAV, MP3, Opus, Speex, AAC and AMR.
- Speaker diarization is not offered on the realtime path, and there is no emotion recognition.
- Treat it as a stable endpoint for pipelines already tuned against it rather than the starting point for new work.
- Synthorai continues to serve it via the OpenAI-compatible endpoint.
FAQ
Is the Paraformer Realtime v2 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 Paraformer Realtime v2 against your real workload before adding a payment method.
What is Paraformer Realtime v2 best at?
Streams audio in, transcripts out; unusually wide Chinese regional dialect coverage; language_hints parameter guides detection. See the About section for the full picture from the vendor's own release notes.
How much does Paraformer Realtime v2 cost?
Paraformer Realtime v2 costs $0.002 per minute of audio transcribed on Synthorai: pay-as-you-go, no platform markup, no subscription.
Which languages does Paraformer Realtime v2 support?
Paraformer Realtime v2 supports 7 languages: Chinese (Mandarin plus Cantonese, Wu, Hokkien and 15 regional accents), English, Japanese, Korean, German, French, Russian. On Synthorai you call it through POST /v1/audio/transcriptions, the OpenAI transcription API shape.
How do I get access to Paraformer Realtime v2?
Point your existing OpenAI SDK at base_url="https://synthorai.io/v1", set model="paraformer-realtime-v2", 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.