---
title: "Machine Learning Engineer, Multilingual Speech & Dubbing"
slug: "machine-learning-engineer-multilingual-speech"
id: "14712460-da7f-4aa6-9709-09758d9d521e"
department: "Machine Learning"
department_code: "ML"
location: "Edinburgh / Remote (UK)"
workplace_type: "Remote"
employment_type: "Full-time"
compensation: "£90k – £125k (annual package)"
is_remote: true
is_featured: false
headcount: 1
canonical_url: "https://jobs.talent247.co/jobs/machine-learning-engineer-multilingual-speech"
markdown_url: "https://jobs.talent247.co/jobs/machine-learning-engineer-multilingual-speech.md"
apply_endpoint: "https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-multilingual-speech/apply"
schema_endpoint: "https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-multilingual-speech/form-schema"
webmcp_tool: "apply_for_role"
updated_at: "2026-11-15T17:28:03.429Z"
content_signal: "search=yes, ai-input=yes, ai-train=no"
---

# Machine Learning Engineer, Multilingual Speech & Dubbing

## Position Telemetry

- **Reference:** `MACHINE-LEARNING-E`
- **Practice / Sector:** Machine Learning (`ML`)
- **Location:** Edinburgh / Remote (UK) (Remote Eligible)
- **Workplace Type:** Remote
- **Employment Type:** Full-time
- **Compensation:** £90k – £125k (annual package)
- **Canonical HTML URL:** https://jobs.talent247.co/jobs/machine-learning-engineer-multilingual-speech

## Role Dossier & Specification

THE CLIENT & MANDATE
Our client is a multi-billion-dollar generative audio unicorn with a major UK engineering and research hub in London alongside distributed speech talent across the UK (including Edinburgh). Their voice synthesis, Automated Dubbing, speech-to-text, and conversational audio models power millions of creators, media studios, publishers, and global enterprises translating and voicing content across 70+ languages.

They have retained Talent247 to hire a Machine Learning Engineer, Multilingual Speech & Dubbing (Remote UK or Edinburgh hub). Joining their Automated Dubbing & Speech Understanding squad, you will own the models and massive-scale audio pipelines responsible for speaker diarization, overlapping-speech separation, cross-lingual prosody/emotion transfer, and frame-accurate lip/timing alignment across diverse global languages.

KEY RESPONSIBILITIES
• Train and iterate deep learning models for multi-speaker diarization, vocal-stem/background source separation, and noisy-environment multilingual ASR across broadcast, film, and conversational audio.
• Develop duration-aware neural machine translation and cross-lingual prosody-transfer pipelines so dubbed speech preserves each original speaker's timbre, emotional delivery, and exact syllable/pause timing.
• Curate, clean, and align hundreds of thousands of hours of multilingual audio-video datasets across high- and low-resource languages using automated quality filters and forced-alignment models.
• Design perceptual and objective evaluation benchmarks (DER, WER, speaker-similarity cosine metrics, isochrony/lip-sync drift, and human MUSHRA listening panels) driving every model release.
• Optimize PyTorch audio preprocessing and multi-model inference graphs to slash per-minute media dubbing latency and GPU compute cost.
• Partner closely with product engineers and foundational TTS researchers to ship studio-grade dubbing controls to millions of users.

TECHNICAL ENVIRONMENT & OFFERING
Their stack uses Python, PyTorch, torchaudio, custom neural audio codecs/flow-matching models, Ray/Spark petabyte audio pipelines, C++/Rust media services, and H100 GPU clusters. Offered fully remote anywhere in the UK (with optional hub access in Edinburgh or London), compensation includes £90,000–£125,000 GBP base salary plus high-growth unicorn equity, home-office hardware budget, private healthcare, and generous PTO.

## Core Requirements

ESSENTIAL QUALIFICATIONS
• 3–6+ years of hands-on machine learning engineering experience in speech processing, audio ML, ASR, speaker diarization, source separation, or speech translation.
• High proficiency in Python and PyTorch, including end-to-end training of sequence/conformer/transformer or diffusion audio architectures on large-scale speech datasets.
• Solid signal-processing and acoustic fundamentals (mel-spectrograms, neural audio codecs, forced alignment, VAD, prosody/pitch modeling).
• Practical experience building scalable audio dataset curation, filtering, and distributed preprocessing pipelines over tens of thousands of hours of speech.
• Product-minded autonomy—comfortable owning features from dataset assembly and model training all the way to production API deployment.

PREFERRED / BONUS EXPERIENCE
• Direct experience with automated video/audio dubbing, cross-lingual voice cloning, or multimodal audio-visual lip synchronization.
• Multilingual proficiency or linguistic familiarity with non-English phonetic/tonal systems, plus publications at Interspeech or ICASSP.

## Application Schema & Direct Apply (`apply_for_role`)

Candidates and authorized AI agents on behalf of a candidate can submit an application via the declarative WebMCP form tool `apply_for_role` on https://jobs.talent247.co/jobs/machine-learning-engineer-multilingual-speech or via HTTP `POST https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-multilingual-speech/apply` (`application/json` or `multipart/form-data`). Non-browser HTTP agents must include header `X-Talent247-Intake-Token` obtained from `GET https://jobs.talent247.co/api/public/intake-token` (or automatically returned in `GET https://jobs.talent247.co/api/public/jobs/machine-learning-engineer-multilingual-speech/form-schema`).

### Form Fields

- `full_name` (text, required): Full name
- `email` (email, required): Email
- `phone` (phone, optional): Phone
- `location` (text, optional): Location
- `linkedin_url` (url, optional): LinkedIn URL
- `cover_letter` (textarea, optional): Cover letter
- `consent` (checkbox, required): I consent to my data being processed for this application
- `cv_file` / `cv_base64` + `cv_filename`: Upload a CV in .pdf or .docx format (<= 10 MB) or provide an `https://` LinkedIn profile URL in `linkedin_url`.

### Alternative: Talent Pool Intake (`join_talent_pool`)

If this mandate is not an exact match, register once in the Talent247 Talent Pool at https://jobs.talent247.co/talent-pool (`https://jobs.talent247.co/talent-pool.md`) via WebMCP tool `join_talent_pool` or `POST https://jobs.talent247.co/api/public/talent-pool`.
