Yandex SpeechKitSpeech-to-speech

speech-realtime-deepseek-v4-flash

JSON

Speech Realtime DeepSeek V4 Flash S2S Alpha streaming, websocket

Experimental Realtime API voice model built on DeepSeek V4 Flash, speaking with general-model voices.

01 Price

$0.78per hour

Cheaper than 49% of 236 priced speech-to-text and realtime models.

Speech-to-text and realtime $ per hour of audio, log scale 236 priced, 280 without a USD price

Speech-to-text and realtime: 236 models with a published USD price per hour of audio, from $0.05 to $20 and beyond, shown on a log scale.

other models this modelAll models by price

Plan or modePrice
outgoing audio$0.60 per hour
input tokens$2.46 per 1M tokens
cached or tool tokens$0.61 per 1M tokens
output tokens$4.10 per 1M tokens

Incoming audio $0.0002163934/s, outgoing audio $0.0001663934/s; tokens $0.002459016 input, $0.000614754 cached/tool, $0.00409836 output per 1,000 tokens (USD excl. VAT).

Pricing source, checked 2026-11-18

02 Languages 2

03 Features 0 of 19 listed

  • Word timestampsnot listed
  • Speaker diarizationnot listed
  • Punctuationnot listed
  • Profanity filternot listed
  • Custom vocabularynot listed
  • Language detectionnot listed
  • Translationnot listed
  • Sentimentnot listed
  • PII redactionnot listed
  • Summarizationnot listed
  • Topic detectionnot listed
  • Emotion controlnot listed
  • Voice cloningnot listed
  • Voice designnot listed
  • SSMLnot listed
  • Phoneme inputnot listed
  • Speed / prosody controlnot listed
  • Style controlnot listed
  • Multi-speakernot listed

"Not listed" means the provider's data doesn't mention it, not that the model lacks it.

04 Voices 28 compatible, browse all

  • No sample published

    Alexander

    male / Russian (RU)

    alexander
  • No sample published

    Amira

    female / Kazakh (KZ)

    amira
  • No sample published

    Anton

    male / Russian (RU)

    anton
  • No sample published

    Dasha

    female / Russian (RU)

    dasha
  • No sample published

    Ermil

    male / Russian (RU)

    ermil
  • No sample published

    Filipp

    male / Russian (RU)

    filipp

05 Get this record

curl https://voicedex.dev/v1/providers/yandex/models/speech-realtime-deepseek-v4-flash

Free, no key needed. Use model id speech-realtime-deepseek-v4-flash with Yandex SpeechKit's own API. API reference.

Esc

↑↓ move↵ open