Migrating Text-to-Speech from Gemini to Sarvam

This guide covers migrating Gemini text-to-speech calls to Sarvam. If you’re building a full voice agent with live, bidirectional audio rather than generating speech from text one call at a time, start from the LiveKit or Pipecat voice agent build guides instead: they aren’t a 1:1 API swap and aren’t covered here.

Quick start

Base URL and auth header change the same way for every Gemini migration, so they’re folded in here rather than in a separate page:

With that, replace the synthesis call and add the required language code:

1
import wave
2
import base64
3
from google import genai
4
5
client = genai.Client(api_key="YOUR_GEMINI_API_KEY")
6
7
interaction = client.interactions.create(
8
    model="gemini-3.1-flash-tts-preview",
9
    input="Welcome to our platform!",
10
    response_format={"type": "audio"},
11
    generation_config={
12
        "speech_config": [{"voice": "Kore"}],
13
    },
14
)
15
16
def wave_file(filename, pcm, channels=1, rate=24000, sample_width=2):
17
    with wave.open(filename, "wb") as wf:
18
        wf.setnchannels(channels)
19
        wf.setsampwidth(sample_width)
20
        wf.setframerate(rate)
21
        wf.writeframes(pcm)
22
23
wave_file("output.wav", base64.b64decode(interaction.output_audio.data))

The REST response is JSON with a base64 audios array, already WAV-formatted; no manual header construction needed. Decode .audios[0] before writing it out; see the REST API guide for the full decode snippet.

The rest of this page covers the full endpoint and parameter mapping, streaming, and common mistakes.

Endpoint map

Parameter concept map

Streaming

What’s different by design

• One flat call, no list to build. speaker and target_language_code are passed directly, rather than nested inside a generation_config.speech_config list, even for a single voice.
• A dedicated TTS endpoint. POST /text-to-speech is purpose-built for speech synthesis, rather than a general interactions endpoint configured to return audio instead of text.
• Ready-to-use WAV output. response.audios is already WAV-formatted; there’s no separate step to wrap raw PCM samples in a WAV header yourself.
• Explicit language targeting. Every Sarvam request carries target_language_code, so pronunciation and prosody are correct by design on every call, especially valuable for Hindi, Hinglish, and other code-mixed Indian-language text.
• Indian language depth is Sarvam’s specific strength. Sarvam covers Hindi plus 9 other Indian languages (Bengali, Tamil, Telugu, Kannada, Malayalam, Marathi, Gujarati, Punjabi, Odia), natively tuned for code-mixed input like Hinglish.

Full example

A small wrapper function, before and after, showing the complete set of changes together:

1
import wave
2
import base64
3
from google import genai
4
5
client = genai.Client(api_key="YOUR_GEMINI_API_KEY")
6
7
def wave_file(filename, pcm, channels=1, rate=24000, sample_width=2):
8
    with wave.open(filename, "wb") as wf:
9
        wf.setnchannels(channels)
10
        wf.setsampwidth(sample_width)
11
        wf.setframerate(rate)
12
        wf.writeframes(pcm)
13
14
def synthesize(text: str, out_path: str) -> None:
15
    interaction = client.interactions.create(
16
        model="gemini-3.1-flash-tts-preview",
17
        input=text,
18
        response_format={"type": "audio"},
19
        generation_config={"speech_config": [{"voice": "Kore"}]},
20
    )
21
    wave_file(out_path, base64.b64decode(interaction.output_audio.data))

Common mistakes

See also

• Migration overview: base URL, auth, and SDK client changes
• Migrating Chat Completion
• Migrating Speech-to-Text
• Which Text-to-Speech API to Use
• TTS Best Practices
• Voices
• Errors & Troubleshooting