
A clear, natural mid-pitched adult female voice in Assamese. Ideal for narration, ads, e-learning and corporate videos.
Bornali, our Assamese voice, plus 800+ library voices that speak it too. Type your text, pick a voice and download at 48 kHz.
Hear the demos before you sign up. Every voice comes with casting notes and documented provenance.

A clear, natural mid-pitched adult female voice in Assamese. Ideal for narration, ads, e-learning and corporate videos.
Today we have one native Assamese voice. And any of our 800+ library voices can read in Assamese too.
Write your text, hear the voices and generate the audio in the studio.
Write the intention in brackets and the voice performs it. Assamese gets the full set: 32 emotions, 18 voice effects and 9 styles. Tags are free.
Native voices sound local. The rest read Assamese clearly, with a hint of their own language: just right for a character from abroad, a brand that wants one voice in every market, or your own cloned voice.
How it works (an example in English and Spanish):
Voiceovers for ads, tutorials and social, without hunting for a voice actor in every market.
Courses in Assamese with one voice from start to finish, and in your other languages too.
Menus, hold messages and announcements in Assamese, as telephony WAV at 8 or 16 kHz.
Articles, documents and websites turned into Assamese audio for people who would rather listen.
One endpoint for all 55+ languages. Pass "language": "as" and pick any voice.
import os, requests res = requests.post( "https://api.voicefinder.io/v1/text-to-speech/09c69440-5d77-471a-9ef6-e8a3a4cbd1a1", # Bornali headers={"x-api-key": os.environ["VOICEFINDER_API_KEY"]}, json={"text": "Your Assamese text", "language": "as", "output_format": "mp3"},)res.raise_for_status()open("speech.mp3", "wb").write(res.content)Each with its own voices, demos and page.
One with voices that do not sound translated. Voicefinder gives you Bornali, our native Assamese voice, plus 800+ library voices, on Voyager 1, our own model: broadcast quality at 48 kHz, direction with emotions and voice effects, and every file marked with an inaudible watermark and Content Credentials (C2PA). You can hear Bornali on this page before you sign up.
Yes. Bornali has Assamese as their main language, with a real local accent. Any other voice in the library can read Assamese too, with a hint of its own language.
Yes. All 800+ Voicefinder voices speak all 55+ languages. Native voices sound local; the rest read Assamese clearly with a hint of their own language, which works well for a character from abroad or for keeping one brand voice across every market. Your own cloned voice can speak Assamese too.
Yes, with the full set of tags. Write the intention in brackets, such as [upbeat], [hushed] or [belly laugh], and the voice performs it: 32 emotions, 18 voice effects like laughs and sighs, and 9 reading styles. Tags are free.
The same as any other language: one credit per character. The Free plan gives you 10,000 credits a month for personal use; commercial use starts with Basic at $12 a month, and yearly billing makes every plan 30% cheaper.
Call POST /v1/text-to-speech/{voice_id} with "language": "as", which sets the language (without it, the model infers it from the text and the voice). The API is on every plan and the audio comes back at 48 kHz, marked.
Free plan for personal use. Commercial use from $12 a month.